The Cipher Brief

I recently returned from commemorating the 82nd Anniversary of D-Day in Normandy—a much anticipated trip since I participated in the 80th Anniversary. I left with a sense of humility and gratitude for having walked the hallowed grounds of Normandy, and with admiration for the Greatest Generation and their sacrifices and actions that changed human history. The connection with Americans, grateful Normans, Europeans who still remember all too well the war, and participants from all over the world who were all there to mark this historic event was remarkable.

Normandy is a place for deep connections. I stood in the door of a vintage C-47 flying low over La Fière Drop Zone listening to the engines roar. Eighty-two years earlier, young paratroopers of the 82nd Airborne jumped into the darkness, many landing in flooded fields or directly into the path of German fire. In the clear light of that June morning in 2026, the hedgerows and pastures looked deceptively serene. What did those young paratroopers see and feel as they jumped into the unknown?

I had the chance to revisit Omaha Beach. The sounds of the waves on that rocky beach can be surreal, an echo of history. If you’ve been there, you remember it. I wondered whether those waves sounded exactly the same on June 6, 1944 as young soldiers and sailors endured withering fire from German positions and refused to be turned back—their sacrifice and refusal to fail now legendary.

In the square of Sainte-Mère-Église, the energetic heart of every Normandy commemoration, I looked up at the church steeple where a mannequin still marks the spot where John Steele hung entangled 82 years ago prior to being captured by the Germans. I imagined the dark sky filled with silk canopies and German ground fire as American paratroopers fought to secure the first town liberated in Normandy and as long-suffering and hopeful Normans awaited the outcome.

Those images are powerful. Normandy is a place that both haunts and inspires, and it invites reflection. It is a powerful reminder of the immeasurable human suffering caused by tyrants and extraordinary cost of freedom. It is hard to visit without connecting with that past; the courage and resilience of the occupied and oppressed, and the audacity and selfless sacrifice of the liberators.

There are so many other images there that make those connections—the American Cemetery with endless rows of white crosses, Pointe du Hoc rugged cliffs and Ranger Monument, Pegasus Bridge, the La Fière causeway and bridge, and many others. Given the scale of D-Day, almost every town, village, beach, road, and bridge witnessed thousands of human stories that shaped the outcome—some well-known, many now faded into history.

The memories and lessons of Normandy transcend time. The threats confronting America today are different from those of 1944, but several enduring lessons remain relevant.

American Leadership Matters

There are things that only America can do, alliances that only America can lead, and geopolitical outcomes that only America can achieve. Eighty-two years ago, the world faced a dangerous shift in the global order. American leadership in WWII was decisive—not only in securing military victory, but in establishing a new world order that has endured for 82 years.

No other country then or now could have led such a global effort, and it cemented America’s emergence as a superpower. Threats in Europe and the Pacific have new faces, but American leadership is as vital today as it was then. Like the authoritarians of that era, Russia and China are actively working to reshape the current world order to their advantage. Both are moving to fill perceived vacuums created by shifts in America’s global posture. Only America can lead the effort to deter them.

If the current world order is to survive and prosper, American leadership on the global stage remains essential.

Alliances Win

America’s leadership role in WWII was crucial, but so were the contributions of its allies and partners. British and Canadian forces played vital roles in D-Day. The French Resistance fought bravely and enabled the work of the American OSS and British SOE. The Soviet Union—then an ally of necessity, now an adversary—played a key role in defeating Nazi forces on the Eastern Front, while other nations—including Poland, Norway, Belgium, and France—made meaningful contributions.

The Allies clearly needed America to enter the war in Europe, but America clearly needed a strong alliance to ensure victory. Global competition has changed—our adversaries are fighting a different war—but reliable partners are still important. Today, alliances are less about planning invasions and more about deterring authoritarian aggression, and building a competitive edge across military, economic, technological, information, and cognitive domains.

America needs its partners. In return, those partners look for a strong and trusted America. Filling that role must remain a national priority.

America’s Story and a National Purpose

Great nations need more than military power to prevail; they also require a compelling national purpose. WWII roused America from its isolationist slumber. The brutal occupation and oppression across Europe, combined with the surprise attack at Pearl Harbor, shattered America’s belief that it could keep major wars at a distance. Powerful themes emerged that were necessary to mobilize an entire nation: “Arsenal of Democracy”, “Four Freedoms”, and the call to defeat fascism.

These ideas fueled a national call to arms, unprecedented mobilization, and prepared America for rationing at home, mass deployments overseas, and the loss of so many Americans who would not return home. America had a compelling national purpose, and it communicated that purpose through a powerful national narrative—we might call that narrative America’s Story. That story included ideals of freedom, democracy, collective security, and the promise of a better postwar order that enabled America to forge and unite an alliance that had previously faltered. That same narrative power remains essential today.

Our adversaries, who seek to reshape the world order, understand the power of America’s story and are working to tell their version of it to their advantage. A persuasive, authentic national narrative—America’s Story—is a national imperative.

Warfare Remains a Human Endeavor

D-Day was a remarkable display of technological innovation. Some technologies were used for the very first time or the first time under such extreme combat conditions. Notable successes, such as the Mulberry Artificial Harbors, the PLUTO undersea fuel pipeline, and improved electronic capabilities that enabled unprecedented coordination of Allied air power played key roles.

Even the highly successful deception operation, Operation Fortitude, relied on technological advancements in jamming and radar deception. In contrast, swimming tanks, gliders, and pathfinder equipment—also rushed into service to achieve surprise and an early advantage—did not meet full operational expectations.

Ultimately, capitalizing on technological advances and compensating for technology failures required human innovation and decisions at the speed of war. In the end, the human dimensions of D-Day—judgment, decision-making in the face of uncertainty, problem solving, detecting and countering deception, courage, and resilience—played a greater role than technology in winning the day and securing victory in Europe.

This is still true today even as AI, robotics, cyber, cognitive tools, and other technologies accelerate the evolution of warfare. A risk is that humans cede decision-making to technology in exchange for greater speed and precision at the expense of judgment, leadership, ingenuity, and moral responsibility. Walking Normandy and seeing the impacts of these human dimensions remind us that while technology, geopolitics, and adversaries change, the human qualities that ultimately shape history do not.

It is these human qualities—not technology itself—that remain America's enduring strategic advantage.

My sincerest thanks to Liberty Jump Team, Corsicana Texas, for an exceptional visit to Normandy for the 82nd Anniversary of D-Day. I’m definitely looking forward to the 83rd.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

What Makes American Patriotism So Different

We are about to close out the 250th year since Tom and the boys stepped brashly onto the world stage and declared with clarity of purpose and universality of meaning our national independence. As proud Americans, we will rightly participate in moments of colorful ostentation and exuberant excess. The 4th of July is flags and fireworks, barbecue and beer, the Air Force’s fabulous Thunderbirds and the United States Marine Corps Band. While certainly some will use the opportunity to cynically enumerate our shortcomings, historical and contemporary, we are, in fact, the longest-standing nation built solely on democratic principles.

Amidst the jubilation, it is probably useful to contemplate the true meaning of American patriotism. Any country can have patriots, but American patriotism is unique. We are not a nation of race or ethnicity, of religion or sectarianism; we do not accept fealty to the absurd notion of inherited “majesty”. We come from all corners of the earth. Nothing makes us inherently better than any other humans. What binds is a set of ideals, the concept of human liberty, and that alone. We do not own this idea; it is human liberty, after all, not American liberty. We are not a people with a set of ideals; we are a set of ideals with a people. We do boldly claim stewardship of these ideals, and, in so doing, we as a nation and as individual citizens accept responsibility for living and acting on them. The conspirators of the Declaration and the framers of the Constitution understood this, and they laid a foundation for a society built on respect for human dignity.

The Constitution itself is a holistic document, imperfect in form, but perfect in meaning. It symbolizes what humans across geography and time yearn for -- a society consciously protecting and celebrating individual freedom while preserving the existence of a nation in which those freedoms may be enjoyed. There are other democratic republics, but there is no other nation like ours. This is the reason people all over the world want to come here. We have what the people of Iran and China want desperately. In contrast, immigration to Russia is the province of the likes of Edward Snowden and Bashar Hafiz al-Asad.

As citizens, we must be attentive to this responsibility we have all accepted, but to put it simply, we could screw this up. As Ronald Reagan famously observed, freedom is always one generation from extinction. That generation does not have to be youngsters convinced of society’s obligation to insulate them from discomfort. It can equally be oldsters who are so lost in narrow-minded rhetoric that they forget the point. American patriotism is not selfish or angry or confrontational or in-your-face or blindly convinced of righteousness. It is the opposite, characterized by humility and generosity, in keeping with our founding ideals.

Whether we were born American or sought and earned our way to citizenship, we have the same burden. Patriotism looks at our tremendous successes as a nation with the same honest eye as our failures, with humble recognition that each contributes to the weight of our obligation to our principles. Patriotism seeks harmony but requires honesty and selflessness. Patriotism calls for deliberate, conscious effort, not just on the 4th of July, but in any action that puts our nation’s credibility on display. We have something no one else has -- no one else has ever had, and the world is watching. As we begin our next quarter of a millennium, America needs patriots as much as ever. Let us seek to be worthy of the stewardship we have been fortunate enough to inherit.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Armenia Has Voted on a Pivotal New Direction, but Moscow is Not Ready to Concede

BLUF: Armenia, a small country of about 3 million people, is located at an important strategic crossroads among Europe, Russia, the Middle East and Central Asia. For centuries, this position has left the landlocked country vulnerable to foreign rule, conflict, irregular warfare tactics, and geopolitical influence. With its 7 June parliamentary election, Armenia has taken an important step in the direction of balancing its foreign policy and playing the role that its geography has positioned it for by bringing these crossroads together. The story is not over however, and Moscow will not let Armenia easily continue down this road. Moscow will continue to pressure Armenia, using all its expertise in irregular warfare, to keep Armenia in its orbit.

Moving Toward the West

Armenian Prime Minister Nikol Pashinyan came to power in May 2018 following a “velvet revolution” that ousted the party that had ruled pro-Moscow Armenia for nearly twenty years. The previous leadership had strong ties to the embattled Nagorno-Karabakh exclave and had made Armenia dependent on Russia for security and its economy.

In the eight years since Pashinyan came to power, he has started peace talks with Azerbaijan, moved the country closer to the West, and cut some ties to Russia. The June 7 parliamentarian election was interpreted as a positive mandate on the direction Pashinyan is taking the country. However, while he won a majority, the pro-Russia opposition held a strong minority position, ensuring that the battle for Armenia’s future was not over.

What the Election Means for a Peace Treaty with Azerbaijan

Last August, the White House brokered a deal between Armenia and Azerbaijan for a ceasefire over ongoing violence in the disputed Nagorno Karabakh territory and set the conditions for the beginning of a peace treaty. As part of the deal, Azerbaijan demanded Armenia approve a new constitution that removed references to Nagorno Karabakh. Changes to the Armenian constitution require two thirds vote in the parliament. Pashinyan failed to secure a two-thirds majority, which means that he cannot change the constitution.

Official preliminary results gave Pashinyan's Civil Contract Party 64 seats in the 105-seat legislature. Two Russian-leaning opposition parties took the remaining seats, while a third appeared to have fallen narrowly short of the 5 percent barrier for entry into parliament pending a recount.

This was the first election since Armenia's military defeat to Azerbaijan in 2023 and a key part of Pashinyan's pitch was what he called "real Armenia," meaning accepting the country's current borders and improving relations with neighbors that have traditionally been hostile -- namely Azerbaijan, but also its patron, Turkey. During the last year, Pashinyan has reached out to the European Union and Turkey, along with the US, to strengthen ties and balance Armenia’s once Russia-focused foreign policy.

In May, dozens of European leaders and representatives of key EU institutions traveled to Yerevan for the first-ever meeting of the European Political Community (EPC) in the South Caucasus, as well as the very first EU-Armenia summit. This was a strong signal toward the West at a pivotal moment for Armenia.

Moscow

The gradual erosion of Russian influence in Armenia may become one of the most strategically important geopolitical shifts in Eurasia since the collapse of the Soviet Union. Armenia was Moscow’s last reliable foothold in the Caucasus. If Russia continues to lose its dominant position in Yerevan, the consequences are likely to extend far beyond the Caucasus, accelerating a broader decline of Russian influence across Central Asia. Underscoring the importance of Armenia to Moscow, President Vladimir Putin threatened Armenia with a "Ukrainian scenario" if it continues building ties with the EU.

For Russia, the loss of Armenia would mean:

· Collapse of Russian strategic dominance in the South Caucasus;

· Weakened Russian military logistics and intelligence networks in the region;

· Decline of Moscow’s political authority among post-Soviet states;

· Destruction of the image of Russia as a reliable security guarantor;

· Expansion of Turkish and Western influence toward the Caspian region.

Most importantly, it would demonstrate that Moscow can no longer preserve its traditional sphere of influence. This would be a major psychological blow to Moscow who is simultaneously fighting a losing war in Ukraine and growing concerns among the population regarding the economic direction of their country.

Moscow has shown that it is prepared to fight for continued influence in Armenia. During the final stretch of the parliamentary election campaign, it launched cyberattacks and disinformation campaigns. The EU sent specialists to help Armenia counter these threats.

In retaliation for Armenia’s continued outreach to the EU, Russia’s Federal Service for Veterinary and Phytosanitary Surveillance banned imports of potatoes, eggplants, fruits — apples, pears, and quince — and dried fruit from Armenia, effective June 3, 2026.

Russia also barred the transit of those agricultural products through its territory to other Eurasian Economic Union member states, citing “the absence of mechanisms to confirm that quarantine-controlled goods have reached those countries.”

The restrictions carry no end date — they will remain in force “until a corresponding procedure is developed to ensure the safety of shipped goods.” More recently, Moscow has banned the import of Armenian fish.

Moscow also has recalled its Ambassador to Armenia for consultations because of the steps taken by the Armenian leadership on a rapprochement with the European Union. The Russian Foreign Ministry said that these steps undermined cooperation with the Eurasian Economic Union. The Eurasian Economic Union (EEU) is a single market which, besides Armenia, is also made up of Russia, Belarus, Kazakhstan and Kyrgyzstan. Moscow has previously stressed that Armenia cannot be a member of both the EU and the EEU.

US Focus

The US is pulling out all the stops to support Armenia. Secretary of State Marco Rubio arrived in Yerevan the week prior to the parliamentary election in a show of support for Pashinyan and his government. President Trump publicly endorsed Pashinyan and his party.

The Trump administration has been working closely with Pashinyan on a road-and-rail corridor initiative called the Trump Route for International Peace and Prosperity (TRIPP), which would run through Armenia and connect Azerbaijan with its Nakhchivan exclave. Rubio said he took another step in the TRIPP project with the Armenian Foreign Minister Ararat Mirzoyan.

"This agreement marks the biggest step to date on making this historic route a reality, on advancing peace and on increasing prosperity in Armenia and frankly in the region," Rubio said at a signing ceremony at Yerevan airport.

Multi-track Foreign Policy

Pashinyan’s focus on diversifying Armenia’s foreign policy makes sense given Armenia’s geo-location. This way forward gives Armenia the possibility of increased status in global trade and potentially in global strategy. Russia will surely attempt to derail Pashinyan and the US’s plans. The question is whether the West will allow Russia to be successful.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Russia’s $11 Billion Soft Power Gamble

I was sitting in the front seat of a “Yandex Taxi” in Moscow in the summer of 2018, dropping off some friends after we had drinks together at a local bar. As the driver, a Russian male in his 50’s, maneuvered through traffic, he asked me if the couple were Americans? “Yes. They are Americans.” “How about you?” he asked. “I am also an American.” Hearing this, the driver continued “You know, 10 years ago I started a small company in Moscow and was doing pretty well. Business was not bad and I enjoyed being an entrepreneur. But then after 2014, when foreign sanctions were placed on Russia my business started to suffer. The Government here claims that sanctions actually ‘help’ our economy, but not in my case. Two years ago I had to shut down my business completely and now I’m driving a taxi to make money to support my wife and daughter.”

Hearing his story, I wondered if this man blamed the U.S. for the sanctions and their impact on his life. “I’m sorry to hear about your troubles” I said. The driver must have understood what I was wondering, and he responded “Don’t misunderstand me. I do not blame you or your country for my problems. I don’t blame the U.S. for the sanctions. No. I blame our “great leader”. Vladimir Vladimirovich Putin. He is the reason for the sanctions. He decided to invade Ukraine in 2014 and, as a result, we were hit with sanctions. And for what reason? Why? Because it would make my life better? Is my family's life better? No. Because it pleased his ego. Because he wanted to feel important and get attention. No. I don’t blame anyone but our own President and his selfish ego.”

This conversation took place during the 2018 World Cup, which was hosted by Russia. Speaking about that event, the driver continued “look around? Look how much money Putin has put into hosting the World Cup. Billions of USD. Our money. The people’s money? Building a new stadium. Re-paving roads, cleaning up cities and refurbishing airports. Teaching police officers and bus drivers how to speak some English and translating street signs and metro signs into English to accumulate the thousands of foreign tourists coming to attend the World Cup. It is all a big show for Putin. An opportunity for him to show off to the world. But will any of this benefit us, the Russian people? Will we see any return on our investment? No. I doubt it. If there is any profit from all of the tourism, it will go into the pockets of Putin and his friends. My family and I will never see any benefit from hosting the World Cup in Russia.”

Last night, watching South Korea play Mexico in the 2026 World Cup, I was reminded of that conversation. If, by any chance, the driver was wrong in 2018 and Russia saw some benefit from the influx of tourists that visited Russia in the summer of 2018, there is little chance that the benefits continued after February 2022, when Putin decided to up the ante and expand his aggression against Ukraine. That disastrous decision has not only resulted in the humiliation of the Russian Armed Forces, deaths of hundreds of thousands of Russians, crippling of the Russian economy but it also resulted in the increasing isolation of Russia. If hosting the World Cup in 2018 was meant to present Russia as a tourist destination for foreigners, how many tourists would dare visit Russia today?

According to some estimates, the Kremlin spent over 11 billion USD to prepare Russia to host the World Cup in 2018. Cities like Moscow and St Petersburg were given makeovers. Other Russian cities that hosted matches had to build new stadiums, and Moscow had to invest billions to build or modernize infrastructure. All part of a Kremlin campaign to present Russia to the world as a modern and developed country. Classic soft power strategy.

At the time, the strategy appeared to work. Hundreds of thousands of World Cup fans poured into Russia. Downtown Moscow was filled with foreign tourists. Despite the political tensions between the U.S. and Russia at the time, and the fact that the U.S. did not even qualify for the World Cup in 2018, Americans made up the largest number of foreign visitors to Russia during the World Cup. During that period, many of the visitors came away with a positive impression of Moscow. Of Russia. Of Russians. Many had a great time celebrating and partying. Russia had a great marketing opportunity to present itself to the world.

Unfortunately, my taxi driver was right. The 11 plus billion USD invested by the Kremlin in Putin’s soft power gamble was erased by Putin’s ego. His gambit to assert full control over his western neighbor in February 2022 led to a significant decline of tourism to Russia and a darkening of Russia’s image in the eyes of the world. How many tourists today would seriously consider visiting a country that is accused of committing war crimes against the Ukrainian population? A country run by an oppressive regime that regularly kills its critics and has forced millions of its own citizens to flee abroad to seek refuge. A country that itself is now being regularly targeted by increasingly effective drone strikes in response to Putin’s continued attempts to terrorize Ukraine. As World Cup fans watched Mexico and South Korea play on 18 June, how many would seriously be interested in visiting the Russian capital, which was covered in dark, black toxic fumes rising from a fire caused by a Ukrainian strike against a Russian Oil refinery in Moscow? I suspect not many.

I often think of that Russian taxi driver, who was trying to make a living after his business was forced to close because of Russia’s growing economic problems in 2018. At the time he understood that without Putin’s ego and poor decision making, there would have been no sanctions. Markets would have remained open to Russia goods and businesses. And, like many Russians I knew at the time, he did not blame Washington, Brussels or Kyiv for Russia’s economic problems. He rightly blamed Putin. And I suspect more and more Russians today feel this way – even if they are afraid to openly admit it.

All statements of fact, opinion, or analysis expressed are those of the author and do not reflect the official positions or views of the US Government. Nothing in the contents should be construed as asserting or implying US Government authentication of information or endorsement of the author's views.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

The Space Age Needs New Rules

Space is no longer a place we visit to plant flags. It is where the global economy and national security now live — and our rulebook is nearly sixty years out of date.

For half a century, space was a government project. Nations went there to prove something about science, about engineering, and about national will. The astronauts were public employees, the rockets were public property, and the point of the exercise was as much symbolic as scientific. But that era is over. What replaced it is both a gold rush and an arms race at once, unfolding in the same orbits under rules written before today.

Two forces have remade the frontier almost overnight. First, private companies now do what only superpowers once could: SpaceX, Blue Origin, Rocket Lab, Planet Labs and dozens of others launch, operate, and profit in orbit at a cadence no government program ever matched. Second, that same orbital infrastructure has become indispensable to national defense and therefore a target. In today's world, satellites are the backbone of how countries communicate, navigate, detect threats, and coordinate military operations. GPS, secure communications, missile warning systems, and real‑time intelligence all flow through space assets. If those systems were disrupted or destroyed, a country's ability to defend itself, project power, or even manage basic infrastructure would be severely weakened.

The result is a domain that is simultaneously more commercial, more crowded, and more contested than at any point in human history, and only getting started. Our institutions were built for none of this. It is time to fix that.

From Flags to Markets

The numbers tell the story of a revolution. The global space economy already approaches half a trillion dollars a year, and credible forecasts put it on a path toward $1.8 trillion or more within the next decade. In the United States alone, the sector already contributes $131.8 billion to GDP each year. Investors poured billions into space startups last year alone, and the United States now captures roughly half of all private space funding worldwide. This is no longer a niche of aerospace contractors living on government cost-plus contracts. It is one of the fastest-growing high-technology sectors on Earth, and the world's wealthiest people are racing to own a piece of it.

The engine of this growth is reusability. When SpaceX learned to land and re-fly its rockets, it did to spaceflight what the shipping container did to global trade: it collapsed the cost. Launching a kilogram to low Earth orbit once cost tens of thousands of dollars; today it can be done for a small fraction of that. Cheaper and faster access changes everything downstream. It is why the United States flew more than 200 commercial launches in a single year, which is the highest annual total this century. It is also why a single company, SpaceX, now accounts for the overwhelming majority of the world's commercial launch activity.

What gets launched has changed too. Instead of a handful of exquisite, billion-dollar satellites, operators now deploy thousands of small, coffee can size, mass-produced ones. Starlink blankets the planet with broadband from orbit; Planet Labs images the entire Earth's landmass every single day; Earth-observation firms sell insight on crops, shipping, emissions, and troop movements to anyone willing to pay. Commercial space stations are being built to succeed the aging International Space Station, and lunar logistics is becoming a business rather than a mission. The center of gravity has shifted decisively from the public sector to the private one.

Make no mistake: this is a triumph. Competition has driven costs down, cadence up, and innovation faster than any government program ever could. But a frontier opened by private capital and moving at commercial speed creates problems that markets alone will not solve. This is where the trouble begins, for which we are not ready.

The New High Ground

The same satellites that power our economy also power our military. Precision navigation, secure communications, missile warning, intelligence, and reconnaissance all run through orbit. Modern forces cannot move, see, or shoot without space, and adversaries know it. That dependence has turned what was once a sanctuary into the ultimate high ground, and the competition to control it is now explicit national policy.

The threat is not hypothetical. U.S. intelligence assesses that China and Russia are fielding a full spectrum of counterspace weapons: ground-based missiles that can destroy satellites, jammers and lasers that can blind or disrupt them, and maneuvering "inspector" craft that can shadow and, if ordered, disable other nations' spacecraft. Officials describe reversible attacks such as jamming and sensor dazzling as occurring on a near-daily basis. Russia's pursuit of a nuclear anti-satellite weapon has been called the single greatest threat to the world's entire space architecture, because a nuclear detonation in orbit would not destroy one satellite but cripple whole swaths of low Earth orbit for years. If this happens, enormous economic impact would occur. Under President Trump, the United States has answered with a declared policy of space dominance: the Pentagon has been directed to ensure American supremacy in orbit, and the Space Force is accelerating the deployment of its own counterspace weapons.

The scale of the buildup is staggering. China operated barely a thousand satellites in 2025; defense planners expect that fleet to approach twenty thousand within fifteen years, many of them dedicated to surveillance and targeting. In response, the United States stood up the Space Force, is spending on the order of $40 billion a year on military space, and is racing to make its constellations resilient, harder to find, harder to kill, and quicker to replace. Crucially, it is doing so hand-in-hand with industry: programs that draw on commercial satellite networks in wartime now treat private operators as part of the national defense fabric. This means government and private sectors are becoming more intertwined.

From a threat standpoint, cybersecurity also needs changing to protect satellites and the networks that control them, because modern space systems behave like connected digital infrastructure rather than isolated hardware. Satellites rely on software, radios, ground stations, and cloud‑based control systems that can be hacked, jammed, or spoofed. A successful cyberattack could disrupt GPS, communications, banking timestamps, aviation routing, or even missile warning systems, creating national‑level consequences. The threat is growing as nations target satellites through cyber intrusions and signal interference, and as commercial constellations expand with software‑heavy, rapidly deployed systems that often have uneven security. Yet international space law barely addresses cybersecurity, leaving countries and companies to rely on their own regulations and best practices. In reality, securing space now requires zero‑trust designs, hardened command links, continuous monitoring, and coordinated defense across governments and commercial operators, because whoever controls the software and signals in orbit controls critical power on Earth.

Here is the uncomfortable truth this creates: the line between a commercial satellite and a military one has all but disappeared. The broadband constellation that connects rural households also connects soldiers at the front. The imaging company that monitors deforestation also tracks armored columns. Private firms are now strategic actors whether they intend to be or not, and that raises questions of law, liability, and protection that no commercial contract was written to answer. It also concentrates extraordinary power in very few hands. A single company, SpaceX, already launches most of the world's payloads and operates the largest constellation ever flown; whoever controls orbital slots, spectrum, and launch capacity increasingly decides who reaches space at all. That is both a triumph and a single point of failure: the Western world's access to space now hinges on the choices of one company, and ultimately one person, which is a degree of dependence few governments would tolerate in any other piece of critical infrastructure. A domain meant to be the province of all humankind now runs on infrastructure owned by a handful of firms and the governments that license them.

Rules Written for a Different Era

So, what are the current rules and what do we do about it? The foundation of all space law is the 1967 Outer Space Treaty. It was negotiated when only three nations had ever reached orbit and governments were the only actors imaginable. It is a magnificent agreement for its time as it keeps weapons of mass destruction out of orbit and declares space the province of all humankind. Yet it sets countries up with a problem: the treaty forbids any nation from claiming territory in space, while granting each nations state jurisdiction over the objects it launches. The effect is sovereignty without ownership: states and the companies they license control satellites, orbital slots, and the data they gather, even as no one is accountable for the domain itself. But it was never designed for a sky full of private mega-constellations and dual-use military assets. No binding space treaty has been adopted since 1979. The rulebook, in other words, predates the personal computer!

The gaps are now operational, not academic. The Outer Space Treaty is reinforced by four companion agreements:

the Rescue Agreement, requiring states to assist astronauts in distress and return them safely
the Liability Convention, which sets rules for compensation when space objects cause damage
the Registration Convention, mandating that states register objects they launch; and
the Moon Agreement, which restricts military activity on celestial bodies and calls for an international regime to govern future resource extraction.

Obligations to act with "due regard" for others, and fault-based liability for collisions, were never given concrete definitions, so they are almost impossible to enforce. And when something does go wrong, the harder problem is proof: with thousands of objects maneuvering through the same orbits and attacks that can be quiet and deniable, attributing a collision or a cyber-intrusion to a specific actor is often impossible, and without attribution there can be no accountability. There is no air-traffic-control system for orbit: each operator largely sets its own collision-avoidance rules, even as tens of thousands of satellites crowd the same shells of space. And there is the debris. Anti-satellite weapons tests alone have scattered thousands of trackable fragments into orbit, a large share of which are still up there, each one a bullet circling the planet at orbital velocity.

The danger is a chain reaction: a collision that creates debris, which causes further collisions, until the most valuable orbits become unusable for generations. National regulators are trying to fill the void piecemeal: the United States now requires defunct satellites to be brought down within five years, and a market for active debris removal is emerging. But orbit is a global commons. Unilateral rules cannot govern a domain where one nation's negligence threatens everyone's access, and the major space powers have shown little appetite for a new binding treaty.

And the gaps are not only physical. As orbit fills with sensors, a harder question trails the hardware: who owns what space sees? A satellite's imagery and signals are raw material for the digital economy, yet the rules for them are thin. The data may belong to the company that gathers it, fall under the jurisdiction of the launching state, or concern people and places that had no say in its capture. The United Nations' remote-sensing principles were drafted for a handful of government agencies, not a commercial market in planetary-scale intelligence. In practice, access is decided by who holds the capability and the capital, raising real questions of privacy, equity, and transparency that no current treaty answers.

Nor is the world negotiating as one. Governance itself is fracturing into rival camps. The United States anchors the Artemis Accords, a non-binding framework now signed by 68 nations, while China and Russia lead a competing bloc around their International Lunar Research Station, joined by roughly a dozen states. Beijing and Moscow have also pressed their own weapons-ban treaty at the United Nations, which Washington rejects as unverifiable. Europe, India, Japan, and a widening circle of newer spacefaring nations move between these poles. The deeper danger is not just that the rules are outdated, but that the major powers are quietly writing parallel rulebooks, none of which binds the others — and any regime that excludes China and Russia governs only the orbits that matter least.

What We Should Do

Managing this new frontier does not mean smothering it. The goal is to keep space open, profitable, and peaceful and that requires governance that moves at the speed of the industry it oversees. We suggest these six priorities to guide us.

Build rules of the road for orbit. We have air-traffic control for the skies and maritime law for the seas; orbit needs the same. A civil space-traffic management system built out from the U.S. Office of Space Commerce's nascent TraCSS program and shared internationally — ideally migrating to a neutral international steward over time — should, within this decade, provide authoritative tracking data, collision warnings, and right-of-way conventions that every operator is expected to follow. The same system should also serve accountability: shared, authoritative tracking makes it possible to establish who did what in orbit, so that a collision or an act of interference can be attributed and answered for rather than denied.
Update the treaty without waiting for a new one. A grand replacement for the Outer Space Treaty is politically out of reach, and far slower than events demand. Instead, adopt the model that works for climate and the oceans: a recurring "Conference of the Parties" convened under UN COPUOS to let nations agree on concrete, incremental standards — definitions of "due regard," deorbit timelines, resource-use norms — without the impossibility of formal amendment.
Make debris mitigation enforceable and universal. Deorbit mandates, building on the FCC's five-year rule, design-for-disposal requirements, and a ban on debris-generating weapons tests should be the global baseline, agreed multilaterally rather than exported by one regulator or left to a patchwork of national rules. Fund and incentivize active debris removal now, while the problem is still merely expensive rather than catastrophic.
Formalize the commercial-defense partnership and its limits. If private constellations are now strategic infrastructure, governments owe their operators clear rules: when commercial capacity can be commandeered, how companies are compensated and protected, and what legal status a private satellite has if it is attacked. Resilience should be bought through partnership, not improvised in a crisis.
Lead through alliances, not isolation. No single nation can police orbit, and the spacefaring democracies are stronger setting standards together. Coalitions among the United States and its allies, with the Artemis Accords as the nucleus, should align licensing, data-sharing, and behavioral norms to build a critical mass of responsible actors that newcomers must either join or be measured against. But coalition-building cannot become bloc-building: the rules that matter most — debris, traffic, and no weapons of mass destruction in orbit — are worthless unless they also bind China, Russia, and their partners, which means keeping channels open through the UN even as alliances set the pace.
Set common rules for space data and fair access. Alongside physical traffic, the framework should govern the information satellites collect: who owns it, how privacy and national interests are protected, and on what terms it is shared. And because slots, spectrum, and launch capacity are finite, and already allocated through bodies like the ITU, access to them should stay open enough that capability and capital alone do not decide who benefits from orbit. Access meant for all of humanity should not narrow into a preserve of the few.

Conclusion

We are living through the most consequential change in humanity's relationship with space since the first satellite crossed the sky. The frontier that nations once visited to prove a point is now where they bank, communicate, navigate, and defend themselves and, increasingly, where they may fight. The private sector has given us an extraordinary gift of capability and cost. National security has made that capability indispensable. What we lack is the governance to match.

The choice before us is not whether to embrace this new era (it is already here) but whether we will steward it wisely or let it descend into congestion, debris, and conflict. The decisions we make in the next few years will determine whether low Earth orbit remains a thriving commons or becomes a contested ruin. We built the rockets that opened this frontier. We are fully capable of writing the rules that will keep it open. We should do so now, before the window closes.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

AI Agents Need Accountability That Travels With Them

Most enterprise AI agents today are still being deployed in controlled environments. They sit inside a platform, perform a defined task and operate under the identity and access controls of that environment, however that window is closing. For many security teams, the current state makes the problem feel manageable. If the agent is inside the fence, the thinking goes, it can be governed by the controls already in place.

That view is understandable. Security teams already have more than enough to manage as AI models become more capable. The near-term implications for vulnerability discovery, fraud, social engineering and incident response are real, and they are moving quickly. Against that backdrop, the question of what happens when agents operate across boundaries can feel like a later-stage concern.

It is not. Identity and access controls can govern an agent inside a particular environment. The harder problem is maintaining accountability when the agent begins operating beyond it. Anthropic’s recentZero Trust framework for AI agents is explicit on this point: each agent instance should have a unique, cryptographically rooted identifier that persists through its lifecycle, appears in logs and access requests, and supports authentication, rotation and revocation. That kind of verifiable identity is what makes safe interoperability possible. As agents move between environments, accountability has to move with them, so the audit trail does not end at the boundary where the risk begins.

The limits of local control

The value of agents comes from their ability to act. They are being designed to invoke tools, coordinate work, exchange information and carry out tasks on behalf of people and enterprises. An agent that can only operate inside one tightly controlled environment may be easier to secure, but it will also be limited in what it can accomplish.

That is the tension enterprises now face. The same interoperability that will make agents valuable will also expand the risk surface around them.

Security leaders are right to focus on the risks already in front of them. More capable models are changing the threat environment in ways that matter right now. But model sophistication is only one part of the issue. The other is autonomy. As agents are given more tools, more permissions and more responsibility, the assumptions behind local control will become harder to sustain.

When Interoperability Becomes A Risk

Most agents are not yet moving freely across enterprise boundaries. Many remain narrow, supervised and limited in scope. But there are two reasons the boundary problem cannot wait.

The first is business value. If agents remain fully contained, their usefulness is constrained. Enterprises will look for returns from AI by connecting agents to more workflows, more tools and more partners. They will want agents to coordinate work across the places where business actually happens.

The second is control. Even enterprises that take appropriate precautions may overestimate how reliably agent activity can remain confined over time. Permissions change. Workflows expand. Tools are added. Business teams find new uses for systems once those systems begin producing value. The environment around agents will keep changing, and accountability needs to remain recognizable when it does.

Security teams are already seeing early versions of this problem in how autonomous AI systems interact with the outside world. An agent exposed to untrusted content may receive instructions the user never sees. An agent with broad permissions may take actions in a context its developers did not fully anticipate. An agent connected to internal data and external communication channels may create a path for leakage or misuse. These are practical control problems, and they become harder to manage as agents gain more tools, more permissions and more autonomy.

For CISOs, the pattern should sound familiar. Some of the hardest security problems emerge at the boundaries between systems, vendors and enterprises. Third-party risk and software supply chain incidents have shown how quickly trust assumptions can break down when no single party controls the full path of activity. Agents introduce a new kind of actor into that same environment. They may be delegated by one enterprise, executed through another platform and interact with a third party in the course of completing a task. In those moments, accountability has to travel with the agent rather than remain tied to the environment where it originated.

The cost of fragmented accountability

In that setting, local identity is not enough. An enterprise may be able to identify and monitor an agent inside the platform where it was created. But once the agent acts elsewhere, that identity may not travel cleanly. Another environment may not know which organization stands behind the agent, whether that relationship can be independently verified, or how trust should be adjusted if circumstances change.

This is where fragmentation becomes a practical security problem. If every platform defines agent identity in its own way, enterprises will inherit a patchwork of trust models. Each may work locally. Together, they create friction at best and gaps in accountability at worst.

Security teams could be left translating between local controls just as agents become more autonomous and more operationally important. That is a difficult place to put defenders. When something goes wrong, they need to know what acted, who was responsible and whether the activity can be contained. Those questions should not depend on which platform created the agent or where it happens to be operating at that moment.

A single high-profile failure could also have consequences beyond the immediate incident. If a rogue or misattributed agent causes material harm, the response could put a chill on the broader market. Security reviews could freeze, integrations could stall and product teams could be forced into a defensive posture as enterprises try to determine which agent activity they can trust. That remains a real risk as long as identity is fragmented and ownership cannot be consistently resolved.

That is not a sustainable foundation for enterprise adoption.

The answer is not to stop agent innovation or force every action back through manual review. That would defeat much of the purpose of the technology. Enterprises want agents because they can move work faster, connect processes and reduce the burden on people. Security teams need a way to support that progress without losing the ability to answer basic questions when something goes wrong.

Which organization stands behind the agent? Can that relationship be independently verified? Can its activity be traced across environments? Can trust be adjusted when circumstances change?

The June 2026 White House executive order on advanced AI innovation and security shows that these questions are now a national priority, one that applies equally across government, industry, and critical infrastructure. The only practical and durable solution for enforcing that is through a standard of accountability that is recognized everywhere.

A standard for portable accountability

Open matters. If the accountability layer for agents is defined separately by every major platform, then trust will fragment at the moment the market needs consistency. Enterprises will face the burden of reconciling competing approaches, and security teams will be forced to govern agents through local controls that do not resolve cleanly beyond their own environments.

A neutral trust layer gives the market a better path. Platforms, model developers, cloud providers and enterprise software companies can still innovate above it. But the basic ability to establish ownership and accountability for an agent should not depend on any one proprietary ecosystem. The trust layer has to be common enough to travel, and that means it’s probably one that already exists.

We have seen this pattern before. The internet was able to grow because certain foundational functions were treated as shared infrastructure. The Domain Name System did not solve every security challenge on the internet, and it was never meant to. But it did create a common way to resolve names across a global network without requiring one company to control the applications and services built above it. AI agents now need a similar foundation for accountability and trust.

That work should begin now, while the agent ecosystem is still forming. Waiting until agents are deeply embedded in enterprise workflows will make the problem harder to solve. By then, fragmented trust models may already be built into products, contracts, integrations and operating processes.

The promise of agents is real. So is the risk surface they introduce. The way forward is to build the accountability layer before fragmented trust models become embedded in the systems that agents will ultimately depend on.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

When Hackers Get AI Co-Pilots: Frontier AI and the National Security Clock

Five intelligence services rarely speak with one voice. When they warn the window of vulnerability has narrowed to months, the real question is whether the defenders can move as fast as the threat.

Throughout my years in the intelligence world, I don’t recall a single instance in which the Five Eyes partners jointly issued a public warning, so when they do, the message lies in the act as much as the words. Intelligence agencies guard their assessments and share them sparingly, almost never in the open. So, when the United States, United Kingdom, Canada, Australia, and New Zealand jointly warned on June 22 that frontier AI models capable of serious cyber exploitation are only "months away" from broad availability, the unanimity was itself a clear message. "The timeline is not years, it is months," they wrote.

The warning the Five Eyes partners shared is specific. These are systems that let a non-expert coordinate a complex intrusion (work that until recently required a trained team fluent in reconnaissance, exploitation, and stealth). That capability is moving out of the hands of advanced nation states and into the reach of mid-tier criminal groups and other adversaries. As the barrier to a sophisticated operation fall, the target list grows, and the systems most exposed are the ones a country cannot do without hospitals, water and power utilities, community banks, ports, and the contractors that serve them.

There is one caveat to mention. Outside experts who examined the models argued they do not represent a wholly novel threat, and the agencies concede their core remedy is familiar: fix the basics, patch faster, control identity and access. The fundamentals still decide most outcomes. What has changed is speed and, with speed, potential volume. The vulnerability was always there, and AI simply finds it faster and puts that reach into more hands.

For national security planners, "months" is the word that should capture attention. Strategy assumes time, and much of the architecture protecting critical infrastructure was built for an era when a capable intrusion took a capable organization. AI collapses that assumption. A defensive posture written to last three years can be overtaken before its first review, and the slowest links (legacy systems and sluggish patching) are the points an adversary will reach first.

Washington has begun to respond. Executive Order 14409, signed June 2, is best read as the opening move in a national security framework for frontier AI. It directs the NSA and CISA to benchmark in classified settings when a model's cyber capabilities make it a "covered frontier model," and it asks developers to voluntarily give the government up to 30 days of access to such models before release. It stands up an AI cybersecurity clearinghouse — led by Treasury — to coordinate the discovery and patching of vulnerabilities, and it directs the Justice Department to prosecute those who turn AI against American computer systems. It also pushes to put defensive AI into the hands of the institutions least able to defend themselves: rural hospitals, community banks, and local utilities.

The order is also a move in a broader contest. Representative Andrew Garbarino, who chairs the House Homeland Security Committee, said the same week that China is "months, if not now weeks, away from achieving frontier AI capabilities comparable to those of the United States." Washington has already moved to restrict the export of a leading frontier model on national security grounds. Whoever fields these capabilities first, and whoever sets the terms for evaluating and controlling them, will shape the rules others must live by. That competition runs straight through the private companies that build the models and the critical infrastructure an adversary would target.

All of this points to the real test. If frontier AI can accelerate attacks, it can accelerate defense, and the side that equips its defenders faster holds an advantage. Programs that put defensive AI into the hands of critical-infrastructure operators, such as Anthropic's Project Glasswing and OpenAI's cyber-defense access effort, are early attempts to give defenders a head start in finding and fixing flaws before they are exploited. The harder problem is people. Models do not run themselves, and the expertise to direct them, in a utility control room or a hospital network, is scarce and unevenly spread across exactly the sectors most at risk.

This is where national security and the private sector stop being separate conversations. Most critical infrastructure is privately owned and operated, which means the front line of national defense now runs through companies whose first duty is to investors and shareholders. The operators that can name the AI systems they rely on, assume their adversaries now carry capable co-pilots, and test their defenses against machine-speed intrusion are the ones that will fare best.

All of this argues for a different compact between government and industry, grounded in shared purpose. Major developers, critical-sector operators, and the national security agencies need to engage early and honestly on the most dangerous capabilities, the way Executive Order 14409 suggests. And the country must invest in defensive AI and in the people who wield it, so the defenders of American systems keep pace with their attackers.

I spent decades in the world of intelligence, much of it managing risk where the cost of getting it wrong was measured in much more than money. The warning the Five Eyes issued this month is the kind that professionals will take seriously. The timeline is tight, and the targets are the systems a society runs on. Frontier AI will define the next era of national power, and the open question is whether the defenders get their co-pilots before the attackers’ finish deploying theirs.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

The War Before the War Has Already Begun

There are 65 active state-based conflicts in the world today, according to the Uppsala Conflict Data Program. That is not 65 separate crises. It is 65 living laboratories.

The contest that matters is not understanding any one of them. It is recognizing the 66th — the next emerging theater — while it is still only a collection of weak signals. The war before the war has already begun, and it will be won by whoever learns fastest.

For generations, intelligence organizations competed to collect more information. Tomorrow, they will compete to learn faster. Since every adversary is becoming a learning organization, our advantage must become organizational learning — and organizational learning at this scale requires infrastructure we have not yet built.

That infrastructure includes a Digital Twin Network.

The Network, Not the Twin

The objective is not to build a better digital twin. It is to build a Digital Twin Network capable of recognizing the 66th emerging theater before it becomes obvious.

Imagine a living network of thousands of interconnected digital twins — not only of nation-states, but of terrorist organizations, criminal syndicates, cyber groups, critical infrastructure, financial systems, media ecosystems, shipping networks, supply chains, political movements and emerging technologies. Every important actor, network and system has a continuously evolving twin.

Each twin learns independently. Collectively, they learn exponentially.

The value is not in the individual twins. It is in the conversations among them. Every observation by one twin makes the entire network smarter. A political crisis in Bosnia immediately updates neighboring political, economic and alliance twins. A cyberattack against critical infrastructure causes financial, media, logistics and influence-network twins to reassess their own environments. A new disinformation tactic discovered in one region is instantly tested against every other emerging theater.

The network does not simply share information. It shares learning.

This is the shift that matters: from monitoring individual events to understanding how thousands of interconnected systems evolve together. From storing information to accumulating learning. From asking “What happened yesterday?” to asking “What is becoming more likely tomorrow?”

What the Network Looks Like in Practice

Picture a digital twin of Bosnia, Moldova or the South China Sea that updates every minute. Every political speech, troop movement, satellite image, shipping pattern, cyberattack, financial transaction and social media narrative automatically changes the model. We move from “what happened” to “what is most likely to happen next.”

AI agents do the work, each with a job. One reads every speech. Another tracks every satellite image. Another looks for new alliances. Another measures the speed of narratives. Together they integrate political developments, military movements, economic indicators, migration, social sentiment, infrastructure, weather, cyber activity and media into a single continuously updated model — one that can identify change in seconds, minutes and hours, and simulate the impact of future actions.

The ability to rank the most successful future actions, based on analysis of hundreds of potential outcomes, changes how we think about red teaming in cognitive security. We will be able to build a synthetic example of every adversary of any size, and to simulate every scenario continuously.

It will be on us to feed in the right inputs. What emerges is a global learning graph of active conflicts — every lesson, every pattern, every conflict feeding better insight in real time.

How the Network Learns: Observe, Learn, Adapt

Conflicts are like a staircase: pressure, politics, perception, prosperity, partnerships, posture, provocation. Every conflict climbs the staircase differently. A network that can read that staircase across every theater at once needs three disciplines.

Observe. We are good at collection. We will benefit from a common structure that makes our observations legible to AI. As an example, The Seven Layers of Emerging Theater Intelligence (SETI) gives every twin the same language for evaluating how adversaries evolve before open conflict:

Pressure — Are underlying conditions becoming less stable?

Politics — Are institutions losing the ability to manage that pressure?

Perception — Is someone deliberately shaping how people interpret events?

Prosperity — Are economic tools becoming instruments of competition?

Partnerships — Are actors beginning to choose sides?

Posture — Is capability being positioned?

Provocation — What event could rapidly accelerate escalation?

Learn. The measure of the network is its learning velocity — how quickly it improves after every observation. Every conflict becomes a research dataset where the network continuously asks: Which indicators appeared earliest? Which signals were ignored? Which combinations proved most predictive? Which assumptions proved wrong? Which interventions slowed escalation? Which technologies changed outcomes?

Adapt. The network tracks how media and technology are evolving and how they will change future tactics. Whether it is artificial intelligence, autonomous agents, commercial satellite imagery, cyber capabilities, sensors, recommendation algorithms or open-source techniques, we watch how each one shortens the distance between pressure and politics, perception and partnerships, posture and provocation.

All of it feeds back into the twins. SETI gives the network a common language; learning velocity gives it a scorecard. Together they make the network something fundamentally different from today’s intelligence systems — a living research community that studies all 65 active conflicts every day and asks the same questions of each. Which pressures are increasing? Which partnerships are changing? Which narratives are spreading? Which actors are learning fastest? And, most important, where is the next theater beginning to resemble the early stages of previous conflicts?

The Scale of the Build

This is why the build matters, and why it must begin now. A network worthy of the threat means digital twins for every nation-state adversary, roughly 100 foreign terrorist organizations, 500 major transnational criminal organizations, 300 state-sponsored cyber groups, hundreds or thousands of hacktivists, 600 militias, insurgencies and armed non-state actors, and thousands of influence and disinformation networks.

That represents a good start.

As AI, autonomous agents and eventually quantum computing mature, the scale of continuous learning will expand dramatically. The future of intelligence will belong to organizations that treat every conflict as a learning system, every emerging theater as a research project, and every observation as a chance to improve faster than their adversaries.

The Only Question That Matters

The race is no longer to understand today’s 65 conflicts. It is to recognize the 66th emerging theater before anyone else — while it is still only weak signals.

That is a contest of learning, and learning at that scale cannot be improvised in the moment a crisis arrives. It has to be built in advance. The Digital Twin Network is that build.

The war before the war has already begun. The only question is whether we will have the network in place to see it.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Congress Questions Pentagon Spending—and the Future of Trump’s Battleship

“I’m deeply concerned that the Presidential proposal for $350 billion mandatory funding [to be carried in a reconciliation bill and not an appropriations bill] for defense will have no Appropriations [Committee] input on the enactment. That’s not the right way to fund the Department of Defense, because it took the Department ten months to explain to Congress how they were going to spend the $150 billion in mandatory funding they received last year. It’s unacceptable, and I have no confidence the Department will do a better job responding to us in the future. There’s also no guarantee that a reconciliation bill will pass.”

That was Rep. Betty McCollum (D-Minn.) speaking last Wednesday at the House Appropriations Committee meeting that marked up the Fiscal Year 2027 Defense Appropriations Bill.

Ranking Democrat on the panel’s Defense Subcommittee, McCollum was questioning the Trump administration’s second year of seeking to put a major chunk of proposed defense spending in a reconciliation bill, where it could avoid both pre-passage congressional review and require only a majority vote for Senate approval.

It turned out that McCollum had bipartisan support for her view.

The House Appropriations Committee, in its report on the bill it later approved that day, included several examples of problems caused by using mandatory spending in a reconciliation bill, along with remedies it proposed..

I will discuss them below, along with one other critical issue – problems in U.S. Navy shipbuilding -- that the House committee also raised in its report.

Remember, however, these are just one committee’s suggestions and they still have a way to go to be adopted by the full House and Senate.

One mandatory spending example in the Committee report relates to the controversial F-35 Lightning joint fighter program.

The President’s fiscal year 2027 budget request includes $7 billion in discretionary funding for 32 F–35 aircraft and $10 billion in mandatory funding for 53 F–35s. Additional modernization funds sought for the F-35 program includes $2 billion in discretionary funding and $2.4 billion in mandatory funding.

In its report, the House Appropriations Committee said it “has serious concerns regarding how the Office of Management and Budget (OMB) bifurcated the funding request and questions the rigor that was used to split the request between discretionary and mandatory funding. For example, radars and other critical components were either funded in full on one side of the ledger

or the other, inconsistent with the total flyaway costs for discretionary and mandatory quantities.”

The Committee report continues, “Further, OMB made assumptions on program savings associated with executing a multi-year procurement contract, for which a corresponding legislative proposal has not been submitted, and applied all the savings to the discretionary request. As a result, the discretionary budget request actually procures a quantity of only six aircraft, rather than the 32 it purports to fund.”

Another Committee report example related to more than $43.4 billion for several critical munitions that is included in the $350 billion mandatory package. The committee said, “In many cases entering into MYP (multi-year procurement) contracts will require both discretionary and mandatory funds. The topic of accelerating munitions production has been a priority of the Department and Congress alike, though splitting funding into two funding processes could lead to incongruencies that will not be easily remedied.”

Splitting weapons programs between the discretionary and mandatory funding prevents Congress from considering requests as a whole, the Committee report says, thus preventing “effective oversight and program continuity and also to preserve production lines and commitments to industry partners and allies.”

The report adds that this year the House Committee is only considering the discretionary portion of the request, but will be “working with the [Defense] Department to ensure that budget justification materials submitted for fiscal year 2028 are adequate to evaluate the full-funding profile, regardless of funding mechanism or whether funding was previously enacted or provided in any future reconciliation package.”

The Appropriations panel report also directs attention to problems in the Navy’s shipbuilding program where the President’s fiscal 2027 budget request includes over $60 billion in discretionary funding for the Trump administration’s so-called Golden Fleet Initiative.

As the report puts it, “The Committee remains firm in its conviction that funding alone does not guarantee on-time delivery and is no substitute for sound program management and rigorous oversight. The Committee is concerned that an accelerated pace of investment, absent commensurate accountability, risks repeating the cost growth and schedule slips that have plagued nearly every major shipbuilding program in recent years.”

Getting specific, the report says, “The Committee is particularly troubled that the Navy’s cost-to-complete request for shipbuilding totals $2.6 billion in fiscal year 2027. The cumulative cost of these delays and overruns now rivals the price of the ships themselves, eroding the buying power of every dollar appropriated for new procurement. The Committee believes that the Navy has not consistently demonstrated the ability to identify, report, and correct adverse cost and schedule trends in a timely manner.”

For a remedy, the Committee “directs the Secretary of the Navy to submit a report to the House and Senate Defense Appropriations Subcommittees not later than 90 days after the enactment of this Act, and quarterly thereafter,” on each major shipbuilding program: to include the current delivery schedule, cost-to-complete with drivers of any growth; and actions the Navy has taken or intends to take to recover any schedule and contain cost growth.

The Committee report also directed the Government Accountability Office next year to assess any recurring cost growth and schedule delay across major Navy shipbuilding programs and the adequacy of the Navy’s response to identify and arrest such trends early.

The Committee also took aim at two specific submarine shipbuilding programs, starting with the Columbia-class which is the sea-based leg of the strategic nuclear triad, and the Virginia-class attack submarine.

According to the Committee report, “the lead Columbia-class submarine is delayed by as much as 18 months and that the Virginia-class program is delayed by as much as 42 months,” adding, “Delays of this magnitude present significant risk to strategic deterrence, erode undersea superiority, and degrade long-term operational availability and readiness.”

Because, according to the Committee report, “incremental funding in a constrained industrial environment serves only to introduce further risk,” the panel recommended “full funding for one Columbia-class submarine and two Virginia-class submarines.”

The House Committee report also took aim at the nascent Trump Guided Missile Battleship (BBG(X) program for which the President’s FY 2027 budget seeks $1 billion in advance procurement and $837 million in research and development funds.

The report says, “The Committee notes that the [Trump battleship] program has not finalized ship design, completed a formal analysis of alternatives, or established a stable set of requirements, and that the Congressional Budget Office has estimated the lead ship could cost in excess of $20 billion.”

The report added the Committee has cautioned in the past that “committing funding to construction before achieving design stability and solidifying requirements is a principal cause of the cost growth, schedule delay, and industrial base instability that afflict Navy shipbuilding.”

The Committee report also warned “that BBG(X), as a nuclear-powered surface vessel, will draw on the same finite pool of nuclear-capable shipyard capacity, skilled workforce, reactor components, and supplier base on which the Columbia-class submarine, Virginia-class submarine, and Ford-class aircraft carrier programs depend.”

Given the situation, the Committee said that “introducing a new nuclear surface combatant [the BBG(X)] without careful planning could compound those constraints and place at risk the delivery of [shipbuilding] programs the Committee considers higher priorities for the nuclear-capable industrial base.”

As a result, the Committee requested detailed reports from the Navy Secretary: One that “addresses the validated requirements and key performance parameters for the large surface combatant [BBG(X)]; the status of the analysis of alternatives and ship design, including a design maturity assessment and the criteria the Navy will use to certify design stability prior to any commitment to lead-ship construction.”

And a second report that deals with the “Navy’s strategy to design and construct BBG(X) without interfering with existing nuclear-powered shipbuilding programs,” and also “how the Navy will sequence and resource BBG(X) so as not to jeopardize the delivery schedules of those programs.”

If that were not enough, the Committee also added a section to the actual legislation, Section 8147, which, by law, would limit the Department of the Navy from using funds to contract to build the lead ship of the Trump-class battleship program, BBG(X), until the “Secretary of the Navy certifies to the congressional defense committees that the weapon systems planned for inclusion in such lead ship are at a sufficiently mature technology readiness level.”

In a column last April, I noted some weapons Trump wants to include on BBG(X) are still in development and any design for such a ship was at least two years away. I now repeat what I wrote two months ago, my bet is that none of these Trump-class battleships will ever actually be built.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

America's Empty Counterterrorism Chair

The world's counterterrorism chiefs are meeting in New York this week. We brought a list of demands and not much else.

A plot hatched in a chat room in one country, funded from a second, carried out by a man walking into a crowd in a third – that's the shape of terrorism now. It doesn't stop at borders. It never agreed to.

That's the problem sitting in front of the world's counterterrorism chiefs this week. They're at the U.N. for the first time since 2023, and on Wednesday the General Assembly reopens the global strategy that's held this fight together for twenty years. The question on the table is simple and ugly: who's actually going to do the work?

Washington's answer lately is everyone but us. The new U.S. strategy tells allies to carry more of the load and barely bothers to dress it up – the era of America as the world's cop is over, pick up the slack. There's a fair point buried in there. Allies should pay more and do more in their own backyards.

But you can't order everyone else to step up while you're sliding toward the door. The federal center built to connect the dots between agencies has had no permanent boss since March. Homeland Security hasn't issued a national threat warning since last September. State shut down its CVE and GEC teams – the shops that countered extremist recruiting and foreign propaganda – over the past year. The FBI and Justice Department teams that chase these cases are thinner than they've been in two decades. We're lecturing the world about leadership while quietly dismantling our own.

And the chair we're vacating doesn't stay empty. The U.N.'s counterterrorism work runs almost entirely on donated money, most of it from a few Gulf states. Cut our funding and our attention, and we don't shut the operation down – we hand the pen to whoever's still paying. Their threats become the priorities. Their enemies become the targets. That swap is already underway, one budget cycle at a time.

The timing couldn't be worse, because the threat is spreading, not shrinking – splitting into more groups, in more places, every year. ISIS-K runs plots out of South Asia. Al-Qaeda's Sahel franchise has turned that region into the deadliest killing ground on the planet. Newer names – the Resistance Front in Kashmir, the Majeed Brigade in Balochistan – show how fast a local grudge now becomes an outfit with cash, recruits, and a slick media shop. You don't beat a threat like that by going it alone – and you definitely don't beat it while eyeing the exit.

So where should the allies in that room actually put their weight this week?

Start with the people nobody wants to claim: the captured fighters and their families still languishing in camps years after the caliphate fell. Bring them home, try them, rehabilitate them. It's slow, ugly, and a political grenade – and every year we dodge it, we let the next generation steep in the same poison that made the last one.

Less visible and more useful is the plumbing: shared watchlists, fingerprints, traveler data, the systems that flag a wanted man before he boards a plane. Most countries still can't run it well, and helping them will stop more attacks than any speech from a podium.

None of that touches the cheapest counterterrorism there is – the kid who never gets recruited in the first place. You can't arrest your way out of this, and the prevention programs that reach that kid early are always the first thing cut and the last thing anyone takes credit for.

And the new front: machines. Cheap drones in the hands of groups that used to throw rocks. AI that spits out propaganda in forty languages and finds a lonely teenager faster than any human ever could. The side that masters these tools first wins. Nothing says it'll be us.

Twenty years ago the world decided this fight couldn't be run one country at a time – not out of idealism, but because the math demanded it. Threats cross borders faster than any single government can chase them. The strategy up for review this week has outlived four presidents for one reason: the work got shared instead of dumped.

Walking away now, with the threat splintering and the tools getting sharper, isn't strength – it's a bet that the next attack will be polite enough to stay in someone else's country. It won't. It never has.

The allies are in the room this week. The only question left is whether we lead the table or leave it.

Dexter Ingram is a former senior national security executive who led the State Department's office for Countering Violent Extremism and served as acting director of the 89-nation Global Coalition to Defeat ISIS. He writes the newsletter Dexter Ingram: Declassified.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Don’t Permit Iran to Enrich Uranium

Ideally, Iran should not be permitted to enrich uranium, even at the 3.67% low enriched uranium level, enough for nuclear reactors to generate electricity, not a nuclear explosion. The Treaty on the Non-Proliferation of nuclear weapons (NPT) does not grant an unfettered right to enrich uranium, but even if it did, Iran’s egregious behavior should disqualify them. Indeed, the NPT recognizes the right of non-nuclear-weapons states to use nuclear energy for peaceful purposes.

Iran has approximately 970 lbs. of uranium enriched to 60% purity – enough for 12 nuclear weapons if in a few weeks enriched to 90% -- that could be down blended to 3.67% in a nuclear weapons state, like Russia, China, or the U.S. Iran may demand, but can they be trusted to down blend this uranium -- buried in deep underground hardened facilities -- themselves?

Enriched uranium for peaceful, civilian use is provided to non-nuclear-weapons states by Russia (Rosatom), France (Orano), a British-Dutch-German consortium (Urenco) and China (China National Nuclear Cooperation). This is where Iran can and should acquire their enriched uranium for peaceful, civilian purposes.

The 2015 Joint Comprehensive Plan of Action (JCPOA) was opposed by some who took issue with the sunset clause that limited enrichment to 3.67% purity for 15 years – until 2031. After that, it depended on nuclear monitors gaining the access necessary to ensure Iran was not enriching at the 20% or 60% purity levels, a few weeks from the 90% purity level necessary for nuclear weapons. The JCPOA also limited Iran to use first-generation centrifuges for 10 years – until 2026 – after which, more sophisticated centrifuges could be developed and used to enrich uranium.

But this should not be a contest between the JCPOA and whatever is decided during the upcoming 60 days of nuclear negotiations with Iran. Our focus should be on Iran and how they succeeded in convincing the U.S. and our allies and partners to trust them to enrich uranium at the 3.67% purity level for civilian use only. But Iran was enriching at the 20% and 60% levels, seemingly threatening to go nuclear, despite their stated commitment not to produce or acquire nuclear weapons. It’s now publicly known that prior to 2003, Iran had an active program to produce nuclear weapons.

The International Atomic Energy Agency (IAEA) often reminds us that their monitors in Iran were denied access to suspect non-declared nuclear sites in Iran. IAEA Director General Rafael Grossi has criticized Iran for severely restricting or denying monitors access to its nuclear facilities and losing knowledge of Iran’s nuclear materials, thus unable to verify that their program was exclusively peaceful.

But it’s more than the NPT and Iran’s so-called right to enrich uranium. It’s about the regime’s evil behavior. In 1984, the U.S. designated Iran the leading state sponsor of terrorism, providing funding, weapons, training, and sanctuary to numerous terrorist organizations. Their control of and support to Hezbollah, Hamas and the Palestinian Islamic Jihad, and the Houthis and the thousands of innocent people killed by these cowardly terrorist organizations should not be forgotten.

Iran’s brutal treatment of its own people – the Green Movement in 2009, the Women Life Freedom Movement in 2022, and the 2025-2026 Protests – and the use of lethal force to crush any anti-government dissent should not be forgotten.

This is the Iran we’re dealing with. Indeed, is this the Iran we can trust to enrich uranium at the 3.67% purity level for civilian, peaceful use?

As we prepare for the 60 days of nuclear negotiations, Iran’s theocracy must feel good about the $300 billion they will get for reconstruction and economic development, and the lifting of sanctions to permit them to sell crude oil and diesel on the open market, and according to the Wall Street Journal, the unfreezing of $100 billion for humanitarian use and Qatar’s immediate release of $6 billion. And of course, the lifting of the 2-month U.S. blockade of Iran.

Iran’s supreme leader, Mojtaba Khamenei, said President Donald Trump acted “out of desperation” to secure the agreement with Iran, while Iran acted “out of compassion and goodwill” to reach the agreement.

Iran’s ballistic missile program and support to proxies should follow our scheduled 60 days of nuclear discussions with Iran. These programs are a threat to the region.

Clearly, Iran cannot be trusted to enrich uranium.

The author is a former Director of the National Counterproliferation Center. All statements of fact, opinion or analysis expressed are those of the author and do not reflect the official positions or views of the U.S. government. Nothing in the contents should be construed as asserting or implying U.S. government authentication of information or endorsement of the author’s views.

This article was originally published in The Washington Times and is republished here with permission.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

The AI Bubble and the Growing National Security Problem

The AI bubble is not a capability bubble. It is an expectation bubble. National security leaders are treating AI as a replacement for analysts, engineers, and tradecraft when it is really a volatile acceleration layer that still requires human judgment, security controls, and cost discipline.

The current state of AI is defined by inflated assumptions. Vendors overstate capability, users over-delegate judgment, and policymakers react to controlled demos as if they represent real-world operational power. The Mythos/Fable incident shows how quickly that confusion can become policy: the U.S. government treated access to a commercial model as a national-security transfer, forcing Anthropic to restrict access to its premier systems.

The problem is not that Mythos is too powerful. The problem is that institutions are starting to make decisions as if the marketing copy is reality. These systems are powerful, but they are not independent thinkers.

AI can surface information at extraordinary speed. It can summarize documents, generate code, translate foreign-language material, identify patterns, and automate repetitive tasks — but it cannot create new ground truth. It cannot determine whether a piece of intelligence is reliable, whether a cyber operation is lawful, or whether an analytic conclusion is strategically sound.

This is where the national-security conversation is going wrong. The debate keeps treating model capability as operational capability. They are not the same. A model that can describe a vulnerability is not the same as an operator who can exploit it. A model that can summarize a document is not the same as an analyst who can assess it. The more powerful these systems become, the more dangerous that distinction becomes.

AI does not exercise judgment, understand mission context, or carry accountability. It is an acceleration layer, and in the hands of trained users, it compresses time and expands reach. In the hands of institutions that mistake output for truth, it will accelerate error, overconfidence, and bad policy.

The bubble is bursting, but not because AI failed

The AI bubble is bursting because organizations bought the wrong story. They thought they were buying replacement labor. What they actually bought was an expensive, overconfident junior assistant: impressive in the interview and with first drafts, but unreliable when placed inside workflows that require judgment, context, and accountability.

Despite the rhetoric of AI replacing jobs, companies are starting to confront a harder reality: these systems can accelerate work, but they do not eliminate the need for people who understand the work. The danger is not simply that AI will produce bad output; the danger is that institutions will mistake that output for finished analysis.

AI is not cheap labor

AI is often sold as cheap replacement labor. The reality is much more nuanced: in proactive it is an expensive acceleration layer that still requires human judgment, review, and correction. At Shadow Nexus, we have AI integrated as a portion of our solution, but it is not the capability itself. Using AI in this manner helps us unlock information hidden in data that would be difficult to reach manually. But this has only worked because our tools requires a human to be involved every step of the way – providing course correction and validation.

That's what makes the "fully autonomous" pitch so misleading. The autonomy is really a system that, left unchecked, is prone to make mistakes and inflate costs.

Microsoft researchers recently tested how major frontier models perform in delegated workflows. They found that even frontier models corrupted an average of 25 percent of document content after 20 back-and-forth interactions, while the average across all tested models was about 50 percent degradation. Degradation worsened with larger documents, longer interactions, and distractor files.

The test was simple: give the model a document, ask it to make an edit, then ask it to get back to its original state. A reliable delegate will returns the document close to its original form. Instead, the errors compounded — like making a photocopy of a photocopy until the original slowly disappears.

The problem is further compounded by the constantly changing pricing model. Anthropic's Opus 4.7 tokenizer increased token usage by up to 35 percent (meaning the same text put into Opus 4.6 would require 35% less tokens). Then with the introduction of Fable 5 only a few months later, Anthropic doubled the published token price.

This rapid increase represents a serious procurement problem for corporations and government customers alike. Agencies can budget for seats, licenses, and fixed contracts. It is much harder to budget for agentic workflows that expand unpredictably through context growth, tool calls, retries, failed tasks, and human rework. That is not just sticker shock. It is meter opacity.

The Tradecraft Problem

Cost is only half the problem. Even at a price you can predict, AI introduces a subtler risk: it produces polished mistakes at scale — and in analytic environments, a polished mistake is far more dangerous than an obvious one.

AI hallucination is not just a chatbot problem. It becomes an institutional risk when generated text enters official documents, legal analysis, or intelligence reporting without source-level verification. Recently, Deloitte Australia agreed to partially refund the Australian government after a report it produced was found to contain AI-generated errors, including nonexistent references and fabricated quotes from a federal court judge.

For intelligence work, the analogy is obvious. A hallucinated citation is not a formatting error, it’s a provenance failure – and a hallucinated provenance chain can contaminate judgment, mislead decision-makers, and jeopardize missions. Don’t misunderstand me: This does not mean AI should be kept out of intelligence work. It means the tradecraft needs to evolve.

AI can be a force multiplier when used to accelerate research, translation, link analysis, and other repetitive analytic tasks - but it should not be treated as a replacement analyst. It has no concept of a larger context, which means it can’t understand legal authorities, operational risk, or true mission context. Those responsibilities still (and should always) belong to people. The right model is not “AI instead of analysts,” it is analysts using AI inside workflows. This requires changing the tradecraft to include a completely new way of thinking.

Which lands a government customer in an impossible spot: how do you adopt and rely on a tool that you can neither fully trust nor accurately budget for?

Government Adoption and the Rising China Problem

For both government and commercial users, the obvious response to rising AI costs is to move towards publicly available "open-weight" models. Systems like GLM-5.2 and Qwen-3.7 now rival the most advanced commercial models, improving cost predictability while keeping sensitive workflows inside government-controlled infrastructure. The catch: they're all designed and shipped from China.

That's what makes the recent Anthropic fight so revealing. Earlier this year, the Pentagon reportedly designated U.S.-based Anthropic a supply-chain risk after a dispute over its safeguards and military use of its models — even as China's GLM-5.2 ranks among the top systems on the market, just behind Anthropic's own Fable 5, with Alibaba's Qwen not far behind.

This is the irony the policy debate: government is trying to regulate a technology it doesn't fully understand, and much of that fear is driven by marketing. Fable 5 is powerful — but so are Opus and GPT-5.5. In the hands of a seasoned user, GPT-5.5 does just as much. As with every new technology, the danger isn't the tool. It's the user.

Meanwhile the drift is already underway. Microsoft recently signaled it may leverage China's DeepSeek model, even as the U.S. weighs blacklisting DeepSeek as a supply-chain risk. Assigning a supply chain risk to U.S. companies feels like an overstep when the trends show organizations moving toward models developed and controlled by adversarial nations.

AI is not going away, and no branding fight or access restriction will change that. The United States should treat AI as the new standard tool for analytic and operational work. But that is all it is: a tool. At its best, it's a starting point — a way to draft, accelerate research, and move faster through large volumes of information. That is also where the handoff to a human has to happen.

The Microsoft research and the Deloitte case are the warning. Left alone, generative AI does exactly what it is built to do: generate plausible output, regardless of accuracy. That risk only compounds as agencies look past closed U.S. models toward open-weight systems built by adversaries.

What happens when the model itself has been trained to nudge its answers — quietly, in a direction someone else chooses? Left uncaught, that kind of slow and deliberate data poisoning can corrupt the very work it's meant to support. That is the real supply-chain risk.

The real work should not be choosing which models we're allowed to use — it should be building the judgment to use them, and not mistaking model names for national-security strategy.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

DNI Day Two: Building the Intelligence Community for 2045

Author's Note

In our first paper, DNI Day One: Three Strategic Decisions for National Security Evolution, we identified three challenges confronting the next Director of National Intelligence: enterprise leadership, resource alignment, and strategic competition. This paper focuses on the reforms most likely to improve the Intelligence Community's ability to meet those challenges.

The recommendations that follow are not intended to be exhaustive. Rather, they represent structural changes that could have disproportionate impact across the Intelligence Community: refocusing ODNI on total enterprise leadership, modernizing how intelligence investments are governed, and creating a mechanism to identify and resolve the institutional seams that America's adversaries increasingly exploit.

Reasonable observers may disagree on the specific solutions proposed here. However, the need for reform is becoming increasingly difficult to ignore. Twenty years after the creation of ODNI, the Intelligence Community faces a fundamentally different operating environment shaped by strategic competition, commercial innovation, artificial intelligence, and increasingly integrated threats. The challenge is not whether the nation needs intelligence reform, but where leaders should focus their attention to achieve the greatest enterprise impact.

Recommendation 1: Return ODNI to its Community Management Roots

We do not recommend eliminating the Office of the Director of National Intelligence. The need for a senior intelligence leader responsible for integrating the Intelligence Community remains as important today as it was when leaders such as Sam Nunn and Richard Lugar first advocated for reform in the 1990s. The failures exposed by the attacks of September 11, 2001, demonstrated the limitations of the existing Community Management Staff structure and ultimately led Congress to establish the DNI.

The challenge today is not whether the nation needs a DNI. It is whether ODNI is organized to perform its most important functions and has the correct authorities and oversight functions to truly lead the intelligence community.

The next DNI should undertake a deliberate reevaluation of ODNI’s mission and functions to refocus the organization on its original purpose: enterprise leadership, integration, resource alignment, and community management. Functions that primarily execute intelligence missions, including analysis, collection activities, and intelligence operations, should be performed by agencies whose core mission is operational execution. ODNI should focus on the responsibilities that only an enterprise integrator can perform.

Specifically,

Conduct a 90-day review of all ODNI organizations against a simple standard: Does this function uniquely require an enterprise integrator? The review should be led by a small panel of respected former intelligence leaders and provide recommendations to both the DNI and Congress. This would include specific evaluations of the transfers of NCSC and NCTC, both of which have been periodically discussed.

Transfer mission execution functions that do not require ODNI ownership to the agencies best positioned to perform them.

Reestablish National Intelligence Managers as true enterprise leaders responsible for integrating priorities, collection, analysis, partnerships, workforce planning, and resources.

Rebuild the ODNI as a modern Community Management Staff focused on enterprise integration, technology governance, workforce planning, and resource alignment

Why

ODNI's greatest value is ensuring that the Intelligence Community performs its missions effectively. The DNI should be a leadership and service organization rather than another operational intelligence entity. The Intelligence Community needs a strong enterprise manager capable of aligning priorities, resolving disputes, integrating capabilities, and driving accountability across agencies.

The Intelligence Community's most significant challenges, including artificial intelligence, commercial data integration, workforce modernization, and emerging technology governance, require enterprise leadership. ODNI should lead these efforts.

Recommendation 2: Integrate the Civilian and Defense Intelligence Enterprises and Modernize Intelligence Investment Governance

The next DNI and Secretary of Defense should establish a shared leadership model for integrating the Intelligence Community's civilian and defense intelligence enterprises. As part of this effort, the Under Secretary of Defense for Intelligence and Security (USDI&S) should serve as the Deputy Director of National Intelligence, creating a direct leadership link between the Office of the DNI and the Defense Intelligence Enterprise.

Together, the DNI and Deputy DNI should establish and co-chair an Intelligence Investment Board responsible for enterprise-level investment decisions in areas where national and military intelligence requirements increasingly overlap, including:

Artificial intelligence and advanced analytics;
Commercial data and data acquisition;
Space-based intelligence capabilities;
Cloud infrastructure and computing environments;
Enterprise architectures and interoperability standards;
Open-source intelligence data procurement;
Collection modernization, including human intelligence

The Board should review major intelligence investments through an enterprise lens, identify opportunities to eliminate duplication, establish common standards, and ensure that capabilities serving both national and military missions are developed and funded as integrated priorities.

Over time, Congress should direct a formal review of the National Intelligence Program (NIP) and Military Intelligence Program (MIP) construct with the goal of transitioning from separate budget categories toward a capability-based investment framework that better reflects how intelligence is collected, analyzed, and delivered in the modern operating environment.

Why

The Intelligence Community has made significant progress integrating its civilian intelligence agencies over the past two decades. Less attention, however, has been paid to fully integrating the broader Defense Intelligence Enterprise, including DIA, the military services, Combatant Commands, and Joint Staff intelligence organizations. Collectively, these organizations represent the majority of the Intelligence Community's personnel, collection infrastructure, and operational intelligence capabilities.

The current structure creates an inherent challenge. The DNI is responsible for leading the Intelligence Community, while the Under Secretary of Defense for Intelligence and Security is responsible for oversight, guidance, and policy across the Defense Intelligence Enterprise. Yet, many of the capabilities that will define future intelligence advantage increasingly serve both national and military missions.

Artificial intelligence, commercial data, space-based collection, cloud infrastructure, and advanced analytic platforms do not neatly align with traditional organizational boundaries. They support policymakers, military commanders, intelligence analysts, and operational forces alike. Maintaining separate governance and investment decisions for capabilities that serve both national and military missions risks duplication, slows modernization, and increases costs.

We recognize that proposals to further integrate intelligence governance between ODNI and the Department of Defense may raise concerns about authorities, resources, and departmental equities. This recommendation is not intended to diminish the Department's role in intelligence or transfer operational control away from Defense organizations. Rather, it reflects the reality that much of the nation's intelligence capability already resides within the Defense Intelligence Enterprise and that effective enterprise leadership requires a governance structure that fully incorporates those resources.

Nor is this recommendation fundamentally about transferring budget authority between organizations. Many of the capabilities discussed in this paper already support missions identified as national intelligence responsibilities under Executive Order 12333 while simultaneously enabling military operations. The challenge is not determining who owns the mission. The challenge is ensuring that enterprise investments are prioritized, governed, and integrated in ways that serve both national and military decision-makers.

To support this integration effort, the Intelligence Community should adopt several enterprise principles.

The DNI and Deputy DNI should jointly establish intelligence priorities and investment guidance. As part of this effort, the National Intelligence Priorities Framework should be modernized into an Intelligence Priorities Framework that reflects the reality that many of today's intelligence challenges span both national and departmental missions. Priorities should be organized around mission outcomes and decision advantage rather than institutional ownership.

The Intelligence Community should pursue acquisition reform focused on enterprise outcomes. Today, agencies often procure similar data, software, cloud services, and analytic capabilities through separate contracts, acquisition strategies, and governance processes. The result is unnecessary duplication, higher costs, slower technology adoption, and inconsistent access across the enterprise. The Intelligence Investment Board should promote joint procurement strategies, establish common requirements where practical, and leverage the collective buying power of the national and defense intelligence enterprises. The goal is a more coordinated approach to acquiring capabilities that support shared missions.

The Intelligence Community should treat commercial capabilities as foundational intelligence infrastructure rather than niche enhancements. Commercial space systems, artificial intelligence, cloud computing, advanced analytics, and open-source intelligence capabilities are no longer supplementary tools. They increasingly form a core infrastructure upon which intelligence advantage depends.

Finally, the Intelligence Community should deepen its engagement with industry, academia, and the broader innovation ecosystem through formal executive and technical exchange programs. Despite decades of outreach initiatives, advisory boards, and pilot programs, the government continues to struggle with integrating commercial innovation at the speed of relevance. Time-limited assignments for leaders from industry, academia, and venture-backed technology firms, coupled with opportunities for intelligence professionals to gain experience in the commercial sector, would help close this gap. These exchanges should operate under rigorous ethics and conflict-of-interest safeguards and be tied to specific objectives such as technology adoption, acquisition reform, commercial integration, and workforce modernization.

For the first time in Intelligence Community history, much of the innovation that will determine future intelligence advantage is being driven outside government. The Intelligence Community's governance, investment, acquisition, and talent management models must evolve accordingly.

Recommendation 3: Establish Strategic Competition as a Core Community Management Function

The next DNI should designate strategic competition as a core Community Management responsibility and direct the Deputy DNI to establish a formal enterprise process for identifying, prioritizing, and resolving the intelligence, counterintelligence, information-sharing, technology, and operational seams that America's adversaries routinely exploit.

This recommendation does not require the creation of a new mission center, operational office, or permanent bureaucracy. Rather, it should be accomplished through a realignment of existing ODNI Community Management responsibilities, leveraging the National Intelligence Managers, Mission Integration organizations, and existing interagency coordination mechanisms.

The Deputy DNI should be responsible for leading this effort in close partnership with the Federal Bureau of Investigation and the Chairman of the Joint Chiefs of Staff, reflecting the reality that strategic competition increasingly occurs at the intersection of foreign intelligence, domestic security, counterintelligence, and military operations.

The objective should be straightforward: identify where institutional barriers are preventing the United States from effectively developing strategic competition options and assign responsibility for resolving those barriers.

The Deputy DNI should provide regular updates to the DNI, Director of the FBI, Chairman of the Joint Chiefs of Staff, National Security Advisor, and congressional oversight committees identifying:

The most significant intelligence integration challenges affecting strategic competition.
Progress in resolving identified institutional barriers.
Recommendations requiring policy, resource, legislative, or executive action.
Remaining risks to national security resulting from unresolved seams.

The effort should establish annual priorities focused on key strategic competitors, beginning with China, Russia, and Iran, and develop measurable objectives tied to enterprise integration and operational outcomes.

Measures of effectiveness should include reductions in information-sharing delays, improvements in technology protection, increased intelligence support to operational missions, improved integration across agencies, and demonstrable disruption of adversary activities.

Why

America's strategic competitors already operate as integrated national security enterprises. They do not distinguish between intelligence collection, technology acquisition, economic competition, cyber operations, influence campaigns, espionage, military modernization, and industrial policy. They employ all instruments of national power in a coordinated manner to advance national objectives.

The United States, by contrast, often organizes its responses through institutional, legal, budgetary, and bureaucratic boundaries that were designed for a different era. As a result, adversaries frequently exploit the seams between foreign intelligence and domestic security, intelligence and law enforcement, technology protection and economic policy, cyber defense and traditional intelligence operations, and national and military intelligence activities.

Intelligence support to strategic competition increasingly requires enterprise leadership capable of integrating activities across the Intelligence Community and identifying barriers that no single organization has the authority or perspective to address independently. This responsibility naturally belongs within ODNI's Community Management function. Strategic competition is fundamentally an enterprise integration challenge. It requires leadership capable of aligning priorities, resolving disputes, integrating capabilities, and ensuring that intelligence activities contribute to broader national objectives.

Also, the DNI, working closely with the FBI and the Chairman of the Joint Chiefs of Staff, is uniquely positioned to bridge the foreign-domestic and intelligence-operations divides that adversaries increasingly exploit.

Ultimately, the Intelligence Community must evolve from simply informing policymakers about adversary campaigns to providing options for the US to compete effectively against adversaries. Strategic competition is the defining national security challenge of the twenty-first century and should be treated as a core Community Management responsibility.

The Ultimate Test of ODNI: Measuring Enterprise Success

As the enterprise leader of the Intelligence Community, this new DNI should demonstrate not only value to Congress and the Executive Branch, but also measurable value back to the agencies, departments, and professionals that comprise the Community. The ultimate test of ODNI's effectiveness is not how many directives it issues, how many meetings it convenes, or how many organizations it oversees. It is whether the Intelligence Community operates more effectively because ODNI exists. The ability to demonstrate measurable outcomes across the enterprise may ultimately be the strongest answer to the recurring question that has followed ODNI since its creation: Does the nation need a DNI? We believe that answer is yes, but that value should be visible, measurable, and attributable.

Specifically, the DNI should be able to demonstrate measurable progress in the following areas to share with OMB, Oversight and the Public:

• Enterprise Priority Alignment
Can the Intelligence Community rapidly align collection, analysis, and resources against emerging national security priorities without creating new organizations or duplicative structures?
(Example: Intelligence Community resources were redirected within weeks to support a Taiwan crisis, major cyber incident, or emerging Arctic challenge without establishing a new task force or mission center.)

• Resource Integration
Has the Intelligence Community reduced duplicative investments and improved enterprise decision-making across the National Intelligence Program and Military Intelligence Program?
(Example: Enterprise licensing and coordinated procurement reduced duplicative commercial data purchases by $15 million annually while expanding access to multiple agencies.)

• Technology Adoption
Has the time required to identify, procure, accredit, and operationalize new technologies been reduced?
(Example: The average timeline for deployment of AI-enabled analytic tools decreased from 24 months to less than 12 months.)

• Commercial Integration
Can commercial capabilities, data, and services be incorporated into intelligence missions at the speed of relevance while maintaining security, interoperability, and mission assurance?
(Example: Commercial GEOINT, RF, maritime, and financial data became available through enterprise contracts rather than separate agency purchases.)

• AI Enablement
Has the Intelligence Community successfully integrated artificial intelligence into collection, processing, exploitation, analysis, and dissemination workflows while preserving human oversight and accountability?
(Example: AI-assisted workflows reduced imagery exploitation timelines by 50 percent while maintaining analytic quality standards.)

• Strategic Competition
Can the Intelligence Community identify, expose, disrupt, and provide options to impose costs on adversary campaigns that span intelligence, cyber, economic, military, and informational domains?
(Example: Intelligence support enabled the disruption of a foreign technology acquisition network or exposed a coordinated foreign influence campaign.)

• Warning and Decision Advantage
Are policymakers, military leaders, and operators receiving integrated intelligence faster and in forms that improve decision-making?
(Example: Strategic warning identified adversary military preparations or cyber activity days earlier than historical performance benchmarks.)

Conclusion

Twenty years after the creation of ODNI, the Intelligence Community faces a fundamentally different operating environment than the one that existed following the attacks of September 11, 2001. Strategic competition, artificial intelligence, commercial innovation, and increasingly integrated threats are challenging many of the assumptions that shaped intelligence governance over the past two decades.

The recommendations outlined in this paper are not intended to redesign the Intelligence Community. Rather, they focus on three areas where reform could produce disproportionate enterprise impact: returning ODNI to its Community Management roots, modernizing how intelligence investments are governed across the national and defense intelligence enterprises, and establishing strategic competition as a core Community Management responsibility.

Taken together, these reforms would strengthen the Intelligence Community's ability to compete more effectively against increasingly sophisticated adversaries. Most importantly, the reforms can largely be accomplished through leadership, governance, and organizational discipline rather than the creation of new bureaucracies or increased resources.

The Intelligence Community does not suffer from a lack of talent, authorities, or capability. Its greatest challenge is ensuring that its capabilities are organized and integrated for the environment it faces today rather than the one it inherited twenty years ago. The next DNI has an opportunity to help lead that transition.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Why One Former CIA Executive Never Stopped Playing

When I was three years old, I fell in love with the violin thanks to an unlikely duet between Itzhak Perlman and a grumpy green Muppet. Watching Perlman’s bow dance across the strings on Sesame Street, I was transfixed. I turned to my parents and announced with all the gravity a toddler can muster, “I want to do that.”

After months of pleading, they relented. That decision didn't just set me on a musical path; it quietly forged my philosophy on joy, participation, and what it means to keep a passion alive for the long haul.

I grew up in a large Irish Catholic family I’d lovingly describe as having “no talent and no shame.” We sang off-key with fervor and staged wildly uncoordinated talent shows every Thanksgiving. While my classmates practiced scales with laser focus, eyeing the first chair in elite orchestras, I played jigs because I liked the way they felt. In my high school orchestra, I sat deep in the second violin section, a world away from the concertmaster’s chair. I wasn’t the best player—far from it—but I loved the sound and being part of a group playing something beautiful, together. That love kept me playing long after others burned out chasing perfection.

After college, I drifted away from the rigid lines of classical music and found my way into the warmth of folk. Dimly lit pub sessions replaced formal concert halls. In that world, you have to listen—to the melody, to the room, to the person across from you—and find your place in the collective sound. I didn't have the fastest fingers or the flashiest solos, but I discovered something vital, I was a great second fiddle.

In folk music, "second fiddle" isn’t a demotion; it’s an art form. It is the realm of harmony, texture, and support. It is the work of making a song richer without needing to be the center of gravity. Without sheet music to guide you, you must listen closely, respond in real time, and improvise. It is collaborative and intuitive—and, for me, far more satisfying than any solo.

Somewhere along the way, I also became the person who started bands. Most recently, I founded the DC-based Irish folk group Celtic Underground, but the spark was lit years earlier at Camp Lejeune. While our husbands were stationed there, I convinced three other Marine wives on my street to join me on my deck for some St. Patrick’s Day music. We lived in officer housing in a neighborhood called Paradise Point, so we jokingly called ourselves The Paradise Pints.

Our skills were ramshackle at best, but we ended up anchoring the neighborhood. The community of Marine officers didn't realize they needed an Irish pub band until they had one. Before long, we had generals singing songs about whiskey at the Officers’ Club on Friday nights. It has been more than five years since I moved away, but the band I started on that front porch is still going strong with a whole new crew of Pints.

I didn't start these bands because I had a grand vision or because I was the most talented person in the room. I did it because I wanted to play music with people, for people (and, selfishly, if I started the band, no one could cut me). I set the tone, picked the tunes, and booked the gigs.

Many of the musicians I gathered were extraordinarily talented—often much more so than me—but they were busy adults with full lives. They were parents and professionals who weren’t going to spontaneously join an Irish band unless someone made it easy and welcoming. When invited into something joyful and low-pressure, they said yes. They were happy to improvise through a new song or learn the bodhrán by watching a YouTube video.

Even in the bands I led, I stayed in the supporting parts. I played second fiddle, sang harmonies, and occasionally moonlighted on the tin whistle. I gravitated toward the background not because I lacked ability, but because that’s where I added the most value. Musically, I wasn't the star, but the band existed because I made it exist.

I noticed this pattern repeating in the most unlikely of places: my professional life. In that world, I was a planner, an overachiever, a list-maker. Up until last year, I was a senior executive at the Central Intelligence Agency, doing hard, complex work and running a large office. I had a seat at the table for key national security decisions, the opportunity to travel the world, brief Presidents. For a long time, I thought I was striving to be the first violin—the concertmaster of a very different kind of orchestra.

Yet, as I reflect on that career, I realize I was rarely the smartest person in the room, nor the one with the deepest technical proficiency. I was "good enough" at the core tasks, but I never truly stood out for my expertise alone. I was "just Meredith." But I kept getting promoted.

It made me wonder: did I achieve what I did because I followed the rules and plowed through to-do lists? Or did the success come from the more intangible things? Just as in a folk session, I thrived because I was the "natural glue." I found joy in stressful situations, understood the changing dynamics of a room, and looked for ways to empower colleagues whose strengths differed from my own. I brought in talented people to work for me and then let them shine. I focused on creating a result that was more than the sum of its parts, and I did it without losing the joy or taking myself too seriously.

Now, as a mother of three in my mid-forties, I play the fiddle several times a week. We tune up in living rooms and Irish pubs around Washington, on small stages and at regional folk festivals. Last year, playing the Takoma Park Folk Festival felt like my biggest career highlight of 2025—a metric that would have baffled my ambitious younger self.

This year, our band is recording our first album. We didn't do it because we suddenly felt "ready" or found a surplus of extra time. We did it because we put it on the calendar, booked the studio, and decided to make it happen.

We live in a culture that prizes being first: first place, first chair, first to speak. But many meaningful parts of adult life don’t disappear because we aren’t "good enough" to continue. They disappear because no one makes room for them anymore.

Sometimes I think about the concertmaster from my high school orchestra—the virtuoso who practiced for hours every day. I wonder if he still plays, or if music became just another achievement to measure, another obligation that eventually fell away. I don’t know the answer. But I know why I’m still here.

Music stayed in my life because I chose participation over perfection. I valued continuity and community over virtuosity. I was willing to start something, show up consistently, and take a supporting role if that’s what kept the group going. I was never the flashiest player, but decades later, I’m still here—still finding the joy in the harmony.

Second fiddle, it turns out, was exactly the right place to be.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

What Iran Wants and How It Can Still Fight

U.S. Vice President JD Vance is touting success out of the latest round of talks in Switzerland focused on seeking a permanent end to the war in Iran. But despite his description of a “very, very good day” of negotiations on Sunday, Iran is denying that it has made any new agreements. It’s more of the same inconsistent messaging the world has become accustomed to over the past three months since the U.S. and Israel launched attacks intended to eliminate Iran’s nuclear program.

Since that time, Iran's supreme leader was killed, the Strait of Hormuz was shuttered and the world has witnessed the largest oil-supply disruption on record.

Now, the United States and Iran are working from a 14-point memorandum of understanding that was signed on June 17, meant to bring the conflict to a formal close within the next 60 days. But will it work and what does Iran stand to gain in the interim?

Cipher Brief CEO & Publisher Suzanne Kelly spoke with former National Intelligence Manager for Iran Norman Roule - who travels regularly to the region for meetings with high-level officials - about what Iran really wants, the impact that a slow-moving and fragile negotiation process could have on near-term energy markets and the tools Iran still has left in its arsenal.

Our conversation has been edited for length. You can watch the full conversation on The Cipher Brief’s YouTube channel.

THE INTERVIEW

Kelly: What does Iran need from this current round of talks?

Roule: Tehran has a new government that needs to prove that it is strong, stable and capable of standing up to its adversaries. And that means upfront, that it can't be perceived as weak, or caving to the United States. Like it or not, the new Supreme Leader cannot sign off on a document that makes him look weak. And if you wish to have diplomatic progress, you're going to have to swallow the fact that a document's going to have to look like something the Iranians can leak and put out publicly and say, ‘we achieved this’. And that just gets you in the room.

The second point is that we're talking about issues that are existential for the regime itself, though not existential for Iran. The country's not going to evaporate. But for the regime, its role with militias in the region, revolution, power projection, keeping down the nationwide unrest - it needs financial relief, it needs to be able to sustain its proxies, it needs to be able to push back on Western military presence in the region. It's negotiating on life and death issues, so it's not going to make any fast decisions.

On the Iranian side, their point is, ‘if we give up a nuclear program, we're giving up our leverage forever’. Sanctions relief is something they need now to sustain the government's survival. They’re not going to give up on the proxies in the region and allow Israel to have a victory. These are real issues in their world, and the administration is just stuck with that reality.

Kelly: How are you looking at U.S. - Israel relationship right now and how significant is the Lebanon issue to reaching any kind of final agreement between the U.S. and Iran?

Roule: This is profoundly complicated. It is as complicated and as consequential as the Strait of Hormuz has been in the actual conflict itself. So let's break this into a couple of pieces.

We now have, as has been predicted for almost two decades, a Revolutionary Guard-dominated government in Iran. There's never been any surprise that the government of Iran would move into a military-dominated government. This war did not produce something that wouldn't have happened by any reasonable analyst’s projection. There was never going to be a reform or moderate government in Iran, period. But that type of military government is going to use two types of tools; military and asymmetric tools.

We've destroyed their entire military. They now have only asymmetric tools left, which they have relied upon for their entire history going back to the 1980s. What are Iran's asymmetric tools? Missiles, mines, drones, cyber tools, and proxies.

The previous government did not defend the proxies when they were attacked by Israel, and the proxies were heavily damaged. That would not have been a choice of the Revolutionary Guard, which is now dominating decision making.

One of the first decisions this government communicated when they talked about the war, that was repeated by President Pezeshkian, and repeated by Javad Zarif in his foreign affairs document was, ‘We're standing up for Lebanon’. Now remember, the Lebanese government, the Lebanese president has said, Iran, we want none of you here. In fact, they've tried to throw out the Iranian ambassador, and he won't leave. That's a very strange situation.

The Iranian government is in essence, is saying, ‘We have a role in protecting our proxies in the region’. Lebanese Hezbollah is only one part of this. There will be no difference between Lebanese Hezbollah, Iraqi Hezbollah and the Houthis. So by doing this, Iran is, in essence, showing power projection and its role in sustaining its proxies.

Now here's the problem. The deal was signed by Iran and the United States, but we're holding Israel accountable to live up to this agreement. Now from the Israeli perspective, they're dealing with Iran, a country that does write, "Death to Israel," in perfect Hebrew on its missiles - and they have a very different political and operational paradigm. But in fairness, they have Hezbollah that does shoot into their country. They have thousands of civilians who have had to move, and their position is that they have to defend themselves.

The question becomes, ‘Is there a Goldilocks zone where their defense can be conducted in a way that doesn't upset a diplomatic apple cart on our side?’ And the Iranians, in essence, can then control the entire process by saying, ‘We're not going to cooperate unless Lebanon is part of this’. So what they're trying to do is not only retain power in the region, but they could use this process to push the United States and Israel into friction against each other, and it has been working.

So what does this mean? This means that a Hezbollah captain or a sergeant - can fire some rockets into Tiberias, or some other Israeli city, kill people, which has happened in the last week and the Israelis respond against that position or other positions - and they have a position of non-proportional response to say, ‘You hit us, we'll hit you bigger so you don't do it again’.

The Iranians then say, ‘What are you going to do? The Israelis are killing civilians’, which has happened in Lebanon.

So now there is intense diplomacy behind the scenes. There's criticism from the United States for the first time, in a very loud way, against Israel, and there is tension. And not for the first time. We had former President Ronald Reagan, we had former President George Bush criticize the Israelis in the past. This administration will absolutely defend Israel and ensure it gets the weaponry it needs to defend itself, but you're seeing political tensions that Iran has been able to manipulate.

I would say it's a mistake to allow Iran any voice in Lebanon. And the international community, including the Arab world, has not done enough to say, ‘Iran, you have no role here. Go away.’ And at the same time, because we failed to do that, just as we failed to dominate the Strait of Hormuz early on, we're going to have some terrific consequences that perhaps may even compromise the success of this diplomatic initiative.

Kelly: I do want to focus on the Strait of Hormuz for a few minutes. There are some competing narratives out there in the energy markets on the near-term supply of oil. How are you looking at the near term oil supply given that we don't really have an agreement yet and it's very difficult for anyone to predict when we might?

Roule: We have to break this into pieces. What we've seen in the last few weeks has been an increase in traffic through the Strait of Hormuz rather than through the Oman side, an informal arrangement with the United States and a number of shippers, and that has reduced pressure on oil markets.

The fact that the Chinese have reduced their purchases has also had a significant impact on oil markets. In the wake of this agreement, we've seen a short spike in the amount of shipping and we've seen gas vessels go into the Gulf to reload, which is important. We've seen movement from multiple ships.

However, this is nowhere near enough. You, in essence, have three different dynamics taking place.

First, the Strait of Hormuz remains mined, and it will take some time to demine this, although less time for us to clear a lane and to say that lane is clear. Insurance companies and shippers are going to seek a certificate or some sort of statement by a world navy to say this lane is clear and it's safe, and then to see a number of ships move through it, and that will cause rates to go down.

Right now, there actually aren't that many ships available to move non-state oil through the Strait of Hormuz. So you haven't seen as much oil go out. And then whenever there's a spike in tension, such as we've seen with Lebanon, you actually see shipping drop. So we've seen shipping drop overnight.

Now once oil comes out, the world will see a lot of oil, prices will drop, and we've seen the market do this. There are a couple of problems though.

We've drawn down enormously on our world's stockpiles. If prices are a little high here, they're much higher in Asia. We've had rationing, governments have shut down, factories have shut down some of the processing in these countries. These countries are going to have to rebuild their stockpiles.

So strangely, as the oil increases in its quantity, we may actually see prices go up a bit as they try to absorb this oil and rebuild those stockpiles. Plus this 60-day ceasefire does not look like it's going to be very successful at present, which means you're going to see countries say, ‘I need to build more faster to get those stockpiles up. Right now, let's not put this oil out there.’

So in the short term, prices are going to come down. They’ll stay in the 80s right now, maybe high 70s. In the longer term, you're probably going to see a bump up. As I've said for a while, late June - July is going to be a tense point. A $10-ish premium is probably going to be likely for a while as countries think of security, stockpile requirements and additional pipeline construction.

Once you get into 2027, you start getting into the possibility of a glut. I would be a little careful at that point because, yes, a glut is possible. But this does depend upon China not purchasing a lot more. This does depend upon continued stability and geopolitics. This does depend upon the international community not picking up its purchases and in the United States continuing to produce at a high level.

So maybe in a few sentences, Short term: prices will continue to go down. Medium term: we shouldn't be surprised if there's a bump up because of stockpile replenishment. A glut in '27 is possible, but we should be careful about saying that it's guaranteed.

Kelly: I'm always asking you what the rest of us aren't focusing on - that you are. I'm curious about the Iraqi militias and the attacks on the GCC countries. How are you viewing the importance and the significance of this and what do you think needs to be done to keep monitoring this?

Roule: It's a story that has not received sufficient attention. There were multiple strikes by Iraqi militias on the GCC during this conflict. There were multiple strikes by Iraqi militias on Iraq during this conflict. The Iraqi militias are clearly trained, and to a sufficiently large extent, under the control of the IRGC.

The United States has invited the new Iraqi leader to Washington. He is a compromised candidate so he is more acceptable than the more pro-Iranian candidate in the past, but he is still acceptable to the pro-Iranian camp within Iraq itself. The administration has sanctioned, I think, the deputy Iraqi minister of energy. And they're going to no doubt continue to pressure Iraq to cut and reduce its ties to Iran's energy sphere and to increase ties to the GCC.

For the GCC, they need to build pipelines and energy connections into Europe through Iraq, but they can't do that through territory that's under the political and security threat of Iraqi militias and indirectly - Iran. It's billions of dollars of capital that's at risk and their energy futures. If Iran can cut the Strait, then Iran can cut the pipelines going north.

So you're going to see a lot of diplomatic and political pressure on the Iraqi government that, frankly, the people in Washington and other places are looking at in a very adult fashion. They know he's in a difficult and delicate political position, but he's going to have to make some hard moves as well. We cannot have Iraqi militias launching missiles on UAE, Saudi Arabia, let alone Israel, competing with potentially Houthis and the Iranians.

And I want to pull this thread just a little bit because of the Revolutionary Guard. We are in a situation now where red lines have been erased.

The red lines of the IRGC using all of its asymmetrical tools, missiles, mines, cyber, militias against everybody all at once, that red line has gone away. So the idea that Iraqi militias won't be used in the future, along with Iranian missiles and cyber, against Saudi Arabia again or Kuwait or Bahrain, that red line doesn't exist. So the Gulf cannot tolerate this perennial weakness in its north as well as in the Houthi south as well as in the east. There's just too much instability. It's too much of a contested region. So Washington will probably put a lot more focus on that.

The other area that I would think there needs to be a little more attention on is the data risk within the region itself. The fiber lines that go through the Red Sea and the Strait of Hormuz contain an enormous amount of financial information, not just from the GCC itself to Asia, but it is also European data flowing between Europe and Asia.

And we've seen the Houthis when they damaged a ship, it had an anchor that damaged several lines that cut some of that data flow for a while but the Iranians are now claiming ownership, and it's fallen out of the news but when it comes to the data line management in the strait of Hormuz - only Iranian companies can repair or manage those lines.

This gives them not only a capacity to control the energy flow and the product but the artificial intelligence flow as well, which the GCC sees as its future to Asia and India and Pakistan - I mean, this is the world. So a GCC that has said, ‘Our future is artificial intelligence and not energy,’ Iran has just said, ‘We will control that future’.

Kelly: The Cipher Brief focuses a lot on gray zone operations and a lot of these undersea cables fall squarely into that category. I wonder if we could talk for just a moment about what Iran is most likely to do during this period of time, What are they doing that they're not talking about?

Roule: The Quds Force has never gone away. Whenever anyone talks about something, one of the foolish phrases of Iran's forward defense, you will often hear people talk about something silly like this, Iran doesn't need drones in the hands of Iraqi militias as a defense. It doesn't need to provide missiles to the Houthis to attack Western shipping as a defense. I mean, anything that someone uses to attack could theoretically be a defense, but the Iranians only call it a defense. And that phrase was originally a propaganda point issued by Iran's foreign ministry and then used by Western shills and then gradually built up into some western think tank narratives. But it's a funny phrase. But you're going to see Iran continue to push out on their asymmetric activities because the Quds Force hasn't gone away, and it's pretty much all they have left. And the Quds Force, to a lesser extent - Iran's Ministry of Intelligence - manages their tools.

So you wisely and eloquently talked about gray zone activities. Iran is - far more than China and far more than Russia - the archtypical gray zone actor. These other countries that have non-gray zone tools and are recognized as non-gray zone powers in the world, but all three are revisionist actors in the world - the three great revisionist actors trying to revise their place in the international community. But Iran only has gray zone tools left because we just destroyed all of their conventional military.

So the Quds Force remains. Any sanctions relief, a small portion of that will go there. The question becomes, ‘What are we doing to cut the logistics lines and what is the international community doing?’ And whenever anyone talks about aid to Iran or assistance to Iran or anything like that, it is not unfair to ask, ‘What are we doing to cut that or how are we measuring Iran's capacity to pull back on the Quds Force?’

If you're in the Trump administration right now, your challenge in the talks going forward is to show that the talks are narrow, reversible, measurable in some way - to show that you're not just providing Iran with the liquidity that Iran and critics of the Memorandum of Understanding will argue it gives. And in return, you're getting something back besides the opening of the Strait of Hormuz. And that something will have to be shown to take place in a few weeks because you can't really do that by day 59 and then say on day 59, ‘We're going to war’.

I'll close by saying that this administration reportedly is saying that there are individuals in the regime who are saying, or telling others, ‘We're willing to move in that direction in exchange for a massive amount of money’. Okay, fine. We'll see. But they're going to have to show measurable examples to prove why something that everyone would say has not been possible for 50 years is going to be remotely possible, and that's going to be hard.

Kelly: Let me close, Norm, by asking you the impossible question. Given how difficult it has been until now and given that you have a very good understanding of the agendas of all sides in this conflict and others who are being affected by this, what do you think a realistic expectation that we will see any kind of measurable progress?

Roule: If the Iranians are able to only create tensions between Washington and Tel Aviv, I think it's manageable, but the rhetoric will be magnified by the press and people who dislike the Trump administration and Israel will magnify that. I don't pay as much attention to that as others because of the politics, but there is a good reason to focus on that.

The likelihood of going back into conflict will depend upon perhaps catastrophic events. I worry about a Hezbollah missile landing and having a large number of casualties in Israel. That creates a gravity sink of actions. Or an Israeli attack doing the same thing, and that could create a gravity sink, or behaviors and political actions that just take us back into a new direction.

Again, the Quds Force has not gone away. That logistics line, all we get is one large shipment of weapons going to Yemen, one large shipment and suddenly something happens. But it could be that we're just in a period of new normal where what we're doing right now might be where we are in July and August and September. People may not like it, but we have been in this position for a number of decades, and we're waiting for the rot within the Iranian regime, which remains a dying regime. It's a stale ideology and a dying regime. That rot will continue to erode the foundations of what's happening there.

I’ll close by saying that we shouldn't overlook the tremendous damage that has been done to the regime during this conflict. It has sometimes been wrongly described as tactical success by the regime. That's terribly wrong. You don't destroy this much of a regime and call it a tactical success. The Iranian government is going to have to try to recover from that, and the brave Iranian people may well rise up in coming months. So there are a lot of ‘What ifs’, but where we are is probably the trend line - barring a catastrophic event of some sort.

Ryan Simons was a producer on this report

The Cipher Brief is committed to publishing a range of perspectives on national security issues submitted by deeply experienced national security professionals. Opinions expressed are those of the author and do not represent the views or opinions of The Cipher Brief.
Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.
Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

From Bombing Iran to Negotiating: Trump Explains His Red Line

“I had to stop them [the Iranians] because if they had a nuclear weapon, they would use it. And you want to see bedlam, let them blow up a couple of cities someplace, like they would've blown up Israel. If it weren't for me, Israel would not exist today, because I terminated the Barack Hussein Obama deal, the JCPOA (the 2015 international agreement limiting Iran’s nuclear program), which was a road to a nuclear weapon. They [the Iranians] would've had it five years ago. They would've used it within the first week, in my opinion. And Israel would no longer be with us. Israel would've been gone years ago had I not done that.”

That was President Trump speaking last Friday on The Axios Show, where his interviewer, Marc Caputo, said it was to be a conversation on “power, and how you [Trump] think about it, and how you wield it.”

Over the next 34 minutes, Trump was direct in his views on war, people and negotiations though what he said at times was not totally factual or he left out relevant information. There were also instances where he seemed to be speaking honestly.

Why does Trump say things that many – if not most people – know are wrong? Also, the President, who in the past has sought to portray his role as a man seeking to end wars – and get the Nobel Peace Prize – during the Axios conversation bragged about his various military operations, including the killing of people. Why?

These are things worth discussing because no matter what he says or does in private, it’s what Trump says or does in public that has an impact and affects how people see him, both friends and enemies including foreign leaders.

For example, had Iran developed nuclear weapons, I believe it is highly unlikely they would have quickly used one or more against Israel. That’s because I know, and Iran’s leaders in Tehran certainly know, although it was not widely publicized, that Israel has for decades had a nuclear arsenal of its own, including more than 90 nuclear bombs and warheads. Some of them are on cruise missiles deployed on Israeli submarines, which means they would likely survive a first Iranian strike and potentially be in a place to strike back.

In short, even if Iran had nuclear weapons, I believe even Tehran’s religious leaders would be deterred from using any of them against Israel because they know that Israel could respond with nuclear weapons of their own.

President Trump, I’m sure, knows about Israel’s nuclear arsenal and the theory of mutual deterrence. A country with nuclear weapons has that kind of protection against other countries with such weapons. Think of North Korea, for example.

Back in 2017, Trump threatened North Korean leader Kim Jong Un with “fire and fury and frankly power the likes of which the world has never seen before.” At that time it was doubted that Pyongyang had a nuclear warhead and missile that could reach the U.S. Since then, it’s clear, the North Koreans do have nuclear weapons that can hit American territory and Trump these days speaks of friendship with Kim.

To me, Trump’s repeated claim that his actions against Iran have “saved” Israel are designed to support his own popularity in that country and, at the same time, imply that the radical Iranian religious leaders are crazy enough to use nuclear weapons against a nuclear-armed Israel.

It reminds me that back in the Cold War period, some top Reagan administration officials promoted the idea that the Soviet Union was considering a nuclear “first strike” against the U.S. because the Russians were constructing their Moscow subway system to serve for civil defense sheltering should we Americans strike back with any nuclear weapons we had remaining.

At another point during the Axios interview Trump said, “I destroyed their general. OK.? [Qasem] Soleimani, [who] was the father of the roadside bomb…It was his favorite weapon. And I killed him. And he killed thousands of [American] soldiers [during the Iraq war] and thousands of other people, tens of thousands of other people.”

Trump went on to describe the details of how he directed the U.S. Special Forces and CIA to carry out Soleimani’s January 2020 assassination in Iraq by blowing up a car he was traveling in, even after Israel backed out of assisting in the operation.

“And it was a flawless attack,” Trump said, adding, “Now, that was one of the biggest moments in the history of the Middle East, because he was the most feared man in 100 years…He was a bad guy, but he was smart. He was a very tough general. You know what he was going to do? He was going to blow up five of our military bases. I got him one week ahead of that attack.”

As for more recently in Iran, Trump said, referring to the U.S. and Israeli February 28 attacks, “I killed the Ayatollah (Ali Khamenei). And a number of IRGC (Islamic Revolutionary Guard) officials. And I sadly hurt the other Ayatollah (Mojtaba Khamenei, the son of the former Supreme Leader and now Supreme Leader), who I will tell you, I did not meet him. I did not speak to him…But he's got a certain braveness because he was, he’s badly injured.”

Trump described what led him to stop the bombing that, I believe, has some truth in it.

“If I were hitting them [the Iranians] right now, when you stopped, if we're not going to put boots on the ground, I mean you don't want [U.S. military] boots on the ground, right? If we're not going to put boots on the ground, probably the same people , they go deep into the caves. They're called granite caves. They're very powerful. They go deep and then when we stop, they'll come up and they'll probably be the same leaders. So nothing, okay?”

As Trump himself emphasized, he wants no part in putting American military on the ground in Iran, or really anywhere in any numbers for any extended time period. Despite his recent turn to military operations, he fears having to take responsibility for American troops being killed or wounded.

He also appears to have learned the limitations of just bombing or using missiles.

Trump said, “When I knocked them [the Iranians] out, we knocked them out so powerfully, we would right now have the Hormuz Strait totally closed. It would have mines all over it and it would have missiles flying over billion-dollar ships. And those ships will never sail.”

Then Trump summed up why he stopped bombing Iran and turned to negotiations.

Trump said, “I just looked. [The price of] oil is tumbling. The ships are roaring out of there [that] they want to go home. They want to go home. They're all full with oil. There's a gusher. I mean, we have seven [hundred] or eight hundred ships are leaving, but if I attack them, none of those ships are leaving. The stock market is way up, way, way up. The stock market is up over the last four or five days when it looks like we're going to make a deal. Stock market's up thousands of points. Everybody's richer. Now, would you rather have that or be like some stupid people?”

He then described “hard liners” who were telling him, “Oh, you got to take 'em [bomb Iranians] out yet.” Of which Trump said, “Well, what lemme tell you. And plus, I'm not looking to kill people. I have one primary wish as president, in terms of people. I never want to be the late great Herbert Hoover. So this is the kind of thing that could cause a worldwide depression.”

In his own words, Trump seemed to recognize the Iran war he started had to be ended before it caused, what he apparently feared coming, “a worldwide depression.”

The outcome of the current 60-day cease-fire negotiations is unclear, but I notice the third item of the Memo of Understanding reads, “The United States of America and the Islamic Republic of Iran commit to negotiating and achieving the final deal in maximum 60 days, extendable with mutual consent (emphasis added).

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Cuba’s New Spy Array Raises Concerns for U.S. Security

BLUFF — On 18 June, Center for Strategic and International Studies (CSIS) researchers released a new study that says Cuba has completed construction of a major signals intelligence antenna array at its Bejucal facility near Havana. CSIS says that based on commercial imagery and open source information, this new construction significantly enhances Cuba’s ability to monitor and locate radio transmissions across a large portion of the Western Hemisphere. This is a specialized listening system designed to intercept radio transmissions and pinpoint their geographic origin with high precision. Construction on the antenna field appears complete, and the CSIS team assesses the facility has very likely begun operations.

In 2024, CSIS identified four Cuban sites featuring equipment that could support signals intelligence (SIGINT) collection, including several with possible links to China. Follow-on analysis of two of these sites, conducted in 2025, found major changes underway at one location, while work at the other had largely stalled. New commercial satellite imagery reveals that activity at both sites has continued.

Commercial imagery of the Bejucal site shows newly completed work on a large circularly disposed antenna array (CDAA). The CSIS researchers Matthew Funaiole, Brian Hart, Joseph Bermudez Jr., and Aidan Powers-Riggs say that this site has undergone a major transformation over the past two years. An older linear antenna grid has been replaced with a Circularly Disposed Antenna Array (CDAA), a specialized system designed to determine the direction and origin of radio signals. According to CSIS, this is the largest and most capable Cuban CDAA installation documented to date. Due to its location near Havana, the Bejucal facility is well positioned to observe U.S. naval operations in the Caribbean, military aviation activity across the southeastern United States, and shipping traffic throughout the Gulf of Mexico.

The Bejucal complex occupies a historically significant military site. The surrounding area was used during the 1962 Cuban Missile Crisis, when Soviet nuclear weapons were stationed in Cuba. In recent years, the facility has frequently appeared in public reporting, congressional testimony, and official U.S. statements regarding foreign intelligence activities in Cuba. While CSIS notes that there is no declassified public evidence proving direct Chinese operation of the specific antenna array, U.S. officials have acknowledged that China operates at least three intelligence facilities in Cuba.

Based on official statements, construction patterns, and previous assessments, CSIS researchers believe Bejucal is likely one of those locations. However, the exact operational arrangements and level of foreign involvement remain undisclosed.

PRC SIGINT infrastructure in Cuba would allow the PRC to:

Monitor U.S. military exercises, missile tests, and space operations.
Map electronic profiles of U.S. assets and communications networks.
Potentially disrupt or influence communications in crisis scenarios

Cuba's proximity to the United States has long made the island strategically valuable for signals intelligence collection, and the completion of the Bejucal array reinforces the role that it can play for our adversaries in surveillance operations against the US. It is ironic that Cuba was ramping up its intelligence capabilities against the US at the same time that the Trump administration is imposing punishing economic sanctions on Cuba. In a May 2026 executive order, probably about the time that Cuba was finishing this upgrade, the Trump administration imposed additional sanctions on Cuba and cited the country's hosting of "foreign adversary facilities" targeting sensitive U.S. national security information.

That the Cuban government has responded to the crippling US economic blockade and sanctions with minor economic reforms while moving forward with increased surveillance activities is concerning but not surprising. This would fit with Cuba’s tactics: appear to be compromising while shoring up its ability to counter the US.

The Cuban government has not acknowledged its intelligence partnerships with US adversaries and now more than ever, the Cuban regime is likely arguing the need is great to better understand threats from the US. What is at risk for the US, however, is the potential that our adversaries can collect intelligence from the myriad military commands that are in the listening zones of these radars. This affects our ability to operate globally and with an element of surprise.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

The Deal That Should Not Have Happened

The announcement of a ceasefire between the United States and Iran surprised many observers. The underlying conditions appeared unfavorable to an agreement—and yet a deal emerged. For months, analysts pointed to profound disagreements over Iran's nuclear program, regional influence, sanctions, and ongoing military activity as reasons why the space for agreement was extremely narrow. Now that an MOU is in place, the immediate temptation is to ask whether it will hold—and perhaps it will not—but a more interesting question may be why it happened at all.

Prior to the agreement, we reviewed an ensemble of simulation scenarios examining hundreds of potential pathways in the evolving U.S.-Iran confrontation. The model incorporated historical, political, and behavioral data to explore how different decisions could shape the course of the crisis. Across the simulation set, indirect engagement, deconfliction channels, maritime security arrangements, and mediated diplomacy appeared repeatedly. Durable, comprehensive ceasefire agreements, however, appeared only rarely. Formal negotiated settlements consistently emerged as low-probability outcomes.

Importantly, this result did not emerge from simple rational-actor assumptions alone. The simulations incorporated leadership behavior, competing bureaucratic interests, and cognitive profiles associated with key actors. The consistent finding was that a comprehensive settlement of the underlying disputes was unlikely.

In negotiations where the observable bargaining space appears narrow, agreements generally emerge for one of three reasons. Either the bargaining space was larger than outsiders realized, the preferences of one or more parties shifted, or the purpose of the agreement is being misunderstood.

The recent ceasefire may contain elements of all three.

The first explanation for the U.S.-Iran MOU is hence straightforward: the public may not be seeing the full agreement. History offers many examples of diplomatic arrangements that included private understandings, sequencing commitments, implementation mechanisms, or parallel agreements that were not immediately disclosed. Negotiators frequently leave politically sensitive concessions outside the public text while embedding them in separate channels.

If this explanation is correct, then the apparent contradiction largely disappears. The agreement was possible because the actual bargaining space was larger than outside observers understood.

The second possibility is more subtle.

Analysts often assume that national interests are relatively stable. In reality, what changes quickly in international politics are neither capability nor intent, but priority. States rarely pursue all objectives simultaneously. Leaders constantly reprioritize. Strategic goals that appeared paramount six months ago may become secondary when confronted with new pressures, risks, or opportunities.

Viewed through this lens, the ceasefire may reflect changing intensities of preferences rather than changing interests. For Tehran, avoiding further military degradation may have risen in importance relative to other objectives. For Washington, avoiding a prolonged regional entanglement may have become increasingly valuable as policymakers confront simultaneous challenges across multiple theaters. If preferences shift sufficiently, an agreement can emerge even when public positions remain largely unchanged.

And there is a third explanation—which may be the most important one. The ceasefire may not actually be designed to settle the U.S.-Iranian(-Israeli) conflict.

One of the most striking findings from our simulation work was the repeated emergence of mechanisms designed not to resolve disputes, but to manage them. Across multiple scenarios, actors established maritime deconfliction channels, backchannels, military hotlines, standing contact groups, intermediary-led diplomatic tracks, and crisis-management frameworks designed to prevent accidental escalation without resolving the deeper issues driving the conflict. In other words, these arrangements did not solve the underlying dispute. They created procedures for living with it. This distinction is increasingly important in a world characterized by persistent strategic competition in multiple theatres.

During the Cold War, many consequential diplomatic achievements were not comprehensive settlements. They were negotiated governance mechanisms. Arms-control agreements, crisis hotlines, incident-at-sea protocols, and confidence-building measures did not eliminate strategic rivalry between the United States and the Soviet Union. They reduced the probability that rivalry would spiral into catastrophe.

Viewed from this perspective, the U.S.-Iran ceasefire may be less significant as a durable resolution of the conflict than as an emerging framework for managing it. If so, its apparent fragility is not necessarily evidence of diplomatic failure. The agreement may not solve the dispute, but it may create enough structure to reduce immediate escalation risks while preserving room for continued competition—and perhaps, over time, a broader diplomatic opening.

This interpretation carries implications far beyond the Middle East. Maritime tensions in the South China Sea, security competition in the Taiwan Strait, navigation disputes in the Red Sea, and strategic rivalry across Eastern Europe may increasingly depend on mechanisms designed to prevent further escalation rather than eliminate disagreement. In each case, the challenge is not necessarily achieving consensus. It is creating enough structure to keep competition from becoming uncontrollable.

Too often, we evaluate diplomatic agreements through a binary framework: success or failure, peace or war, settlement or breakdown. Yet some of the most consequential arrangements in international politics occupy a different category altogether. They function as temporary governance systems for managing unresolved disputes.

In such environments, the relevant question is not whether a conflict has been solved. It is whether sufficient structure exists to keep competition from becoming uncontrollable.

That broader perspective may ultimately matter more than the details of the current ceasefire itself. As geopolitical competition intensifies, durable settlements may become rarer. In their place, we may see the growing importance of improvised, often fragile arrangements around maritime chokepoints, contested regions, and strategic flashpoints that function less as peace agreements and more as operating systems for managing persistent rivalry.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Before the IC Trusts AI, It Needs to Prove It Can Assure It

Artificial intelligence is moving quickly into national security work. That is not a future trend. It is already happening in analysis, collection support, cyber defense, logistics, language processing, software development, and mission planning.

The real question is no longer whether AI will be used, it is.

The harder question is whether we can trust it inside mission environments where bad data, weak access controls, poor model governance, or untested automation can create real operational risk.

For years, cybersecurity leaders have been trained to think about systems, networks, endpoints, identity, and data. AI changes that model. It does not replace those risks; it adds a new layer of uncertainty on top of them. An AI system can be technically functional yet unreliable, manipulated, over-permissioned, poorly sourced, or impossible to explain.

That is a problem in any enterprise. In national security, it is a significant mission risk. AI assurance is not just a compliance exercise. It is the discipline of proving that an AI-enabled capability is fit for purpose, secure enough for its environment, monitored after deployment, and governed by people who remain accountable for the outcome.

Most organizations still treat AI adoption as a technology deployment. Buy the tool, issue a policy, run a pilot, brief the results. That approach may work for low-risk productivity use cases. It does not work when AI is connected to sensitive data, operational workflows, classified environments, or decision support. The model is just part of the risk. The larger risk is the infrastructure around it. In a traditional system, we asked: who has access to the data? In an AI-enabled workflow, we also have to ask: what can the model infer, summarize, combine, expose, or act upon once access is granted? A user may not be authorized to see every underlying source in a system, but an AI tool connected to that system can, and may generate a summary that reveals sensitive relationships, operational context, or protected information.

The same is true for retrieval-augmented generation (RAG). RAG can make AI more useful by grounding responses in ‘trusted’ data. However, it can also create a new attack surface if source material is stale, poisoned, poorly labeled, or pulled from repositories with weak access controls. If the retrieval layer is not governed, the model can confidently produce bad answers from bad inputs.

The answer is not to slow-roll AI into irrelevance. The answer is to operationalize assurance. There are five things national security organizations and cleared industry should be doing now.First, inventory AI use cases like mission systems. Leaders need to know what AI capabilities are being used, what data they touch, who can access them, and what decisions or workflows they influence. Shadow AI is not a user behavior problem alone. It is usually a signal that the enterprise has not provided secure, usable options fast enough.

Second, treat data provenance and lineage as core requirements for data management. AI assurance starts before the model ever generates an answer. Organizations need to know where training data, reference data, embeddings, and retrieval sources came from, how that data moved through the environment, how it was transformed, who validated it, who can modify it, and whether those changes are logged. Provenance tells us the origin of the data. Lineage tells us what happened to it along the way. Without regimented data management, the organization cannot confidently assess whether the model’s output is accurate, up to date, authorized, or appropriate for the mission. If the data supply chain is weak, opaque, or poorly governed, the AI output is already questionable.

Third, test AI models against mission-specific use cases. This could include adversarial prompts, poisoned documents, prompt injection, tool misuse, and hallucinated citations and references.

Fourth, monitor after deployment. Models change. Data changes. User behavior changes. Threat actors adapt. Assurance has to be continuous and include logging, drift detection, output review, access monitoring, and clear thresholds for when a tool should be paused, updated, restricted, or removed.

Fifth, keep humans accountable. Humans-in-the-loop should have clear and accountable responsibilities defined. What is the reviewer expected to verify? What decisions can never be fully delegated to the AI tool?

The organizations that get this right will be the ones that build disciplined AI operating models. They will have clear use cases, controlled data access, measurable evaluations, audit trails, and documented risk ownership.

AI is becoming one of the most important force multipliers in national security and economic competition. It has the potential to narrow gaps between larger and smaller countries, established and emerging companies, and well-resourced and resource-constrained organizations. Capabilities that once required large teams, specialized infrastructure, or years of institutional advantage are becoming more accessible through AI-enabled tools. That is why assurance matters. For the Intelligence Community and the national security industrial base, AI assurance should become a core discipline. Before we scale AI into mission operations, we need to prove we can govern it, test it, monitor it, and explain when it should not be trusted.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Learning Velocity: The Next Strategic Advantage

Entrepreneurs realize that speed compresses learning. They know some of their initial assumptions will be wrong or only partially right, so going fast allows them to test ideas and move from opinions to evidence to products before they run out of cash or time.

A startup that spends years to get to market may end up solving the wrong problem. The startup that launches in weeks or months figures out what their audience cares about, what is irrelevant and they evolve their approach in real-time. Test, learn, adapt and repeat the cycle.

The most successful entrepreneurs know that the original idea is often less important than the rate at which they improve it.

This thinking has existed for decades. Now, AI enters stage left.

The Importance of Learning Velocity

We are entering an era where learning velocity is reducing the time between a question and an answer, an idea and a test and a mistake and its correction.

Learning Velocity is the speed at which an individual, organization, or nation converts information into capability, decisions, and action. Increasingly, competitive advantage is determined not by who possesses the most information, but by who learns from it fastest.

AI changes the speed and quality of the learning cycle.

Can we understand if a disinformation attack is starting before it succeeds?

Can we routinely build synthetic adversary simulations worldwide to improve our preparation?

Can we see 10,000 new accounts emerge in a few weeks to amplify identical narratives across five platforms in five languages and know exactly who is driving it?

Can we identify the next TikTok or Discord well before it gains altitude?

Adversaries realize the pace at which they learn is often as important as what they learn if they are to win the race to market. From Silicon Valley to Bekaa Valley, the learnings are the same. It’s all a matter of how it is applied.

Are We Ready?

Throughout history, strategic advantage has often belonged to those who learned faster than their adversaries. U.S. military doctrine embraced the OODA Loop—Observe, Orient, Decide, Act—as a model for accelerating decision cycles faster than opponents.

Times have changed. Bad actors are compressing learning cycles from months to days and increasingly from days to hours. The public and private sector too often operates on models of learning that are designed for a slower century.

Their advantage is not necessarily better technology, more information, or greater resources. Their advantage is that they learn faster.

Let’s imagine how they learn for a minute.

Imagine the Fastest-Growing University in the World

It has no campus, no admissions office, no tuition, and no accreditation.

Its students come from every country. Its curriculum updates daily. Its teaching assistants are powered by artificial intelligence.

Its graduates include cybercriminals, fraudsters, influence operators, hostile intelligence services, extremist groups, and increasingly, highly capable lone actors.

Welcome to the New University for Bad Actors

Freshman Year: Open-Source Intelligence

Freshmen begin with open-source intelligence 101.

Their textbooks include LinkedIn, Google Earth, company websites, SEC filings, public procurement records, satellite imagery, and social media.

Their first assignment is simple: learn everything possible about a target without ever touching its network.

They learn how to map executive organizational charts, vendor relationships, facility layouts, and employee behavior patterns almost entirely from publicly available information. Whether or not they put on the “freshman 15”, they will learn how to use commercial satellite imagery, drone footage, all forms of digital media, ship, aircraft and supply chain tracking data, patent filings, geolocation tools and more.

Sophomore Year: AI-Assisted Learning

This is not followed by a sophomore slump. Rather, they rapidly move into AI-assisted learning.

Their professors are ChatGPT, Claude, Gemini, Perplexity, Grok, DeepSeek, and other large language models.

What once required months of research can now be accomplished in hours.

Generative AI enables bad actors to create deepfake videos, cloned voices, fabricated identities, realistic images, and highly personalized communications at scale. AI agents can conduct individualized outreach to thousands of targets simultaneously, making social engineering attacks more believable and more effective.

They learn how people form trust. Why a narrative resonates or fails with an audience. How to choose the right topics, build a plan and create polarization. And they start to see where trust is weak and where attacks may be more successful.

They are becoming better students of human behavior.

Junior and Senior Year: Laboratories and Code

Juniors enter the laboratories ready for one of the most important shifts in intelligence and strategic competition.

Hugging Face, GitHub, ArXiv, Discord developer communities, Papers with Code, Stack Overflow, Reddit, Kaggle, and open-source communities become classrooms where they learn how innovation occurs and where the world is going.

They learn which AI capabilities have the most promise in the next year. They can see which technologies are gaining acceptance and which ones are losing steam. Perhaps most important, they witness which workforce skills will be most valuable. They know the performance benchmarks, which datasets matter and where the problems lie.

These students realize they can see which technologies have promise years before they become mainstream.

They are also keenly aware of the speed of innovation. They can read the signals of repository growth, model releases, publication speed and more.

The graduation ceremony is brief. No champagne, just more learning ahead.

Graduate Studies: Shared Learning Networks

Graduate students enter encrypted networks, private forums, and invite-only communities where ideas, techniques, and lessons learned are exchanged among peers. Successful attacks are dissected and analyzed. Indicators of compromise, response timelines, victim profiles, and operational mistakes are openly discussed so that future attacks become more effective.

Every successful operation becomes a lesson for the next one.

Some students have more to learn.

Doctoral Research: Data as a Weapon

Doctoral candidates study commercial data ecosystems, public records, and digital surveillance.

They discover that some of the most valuable information does not need to be stolen. It can be purchased.

Post-doctoral researchers venture into dark web marketplaces and specialized forums where knowledge, tools, services, and expertise are exchanged at extraordinary speed. Personal information, new malware, breached corporate data, criminal service marketplaces, critical infrastructure targeting and more.

The dark web and shared learning networks provide that competitive advantage – information others don’t have.

Why Are They Learning Faster Than We Are?

Bad actors are not necessarily learning more than we are. The real difference is how fast they learn.

When resources are constrained and survival depends on adaptation, speed of learning becomes a competitive weapon. The most successful entrepreneurs build companies with an idea, often inadequate finance resources and a ticking clock showing they could run out of time.

Speed is not only an asset, it is about survival of the idea.

A New Era Deserves a New Model

The challenge for many established institutions is not a lack of intelligence, talent, or resources. It is resistance to change.

We don’t like failed pilots, so people stop experimenting. Our budgets are annual. Our feedback loops are slow. We may learn quickly, but we act slow.

Behavioral economists refer to this as loss aversion. Humans experience the pain of losing existing processes, authority, expertise, and familiar ways of working more intensely than the potential gains associated with adopting new approaches.

The reality is learning faster than adversaries and adapting faster will be our advantage to create and sustain.

Learning Velocity will become our long-term strategic advantage, which means we need a practical way to evaluate our progress.

The model is defined as the speed at which an individual, organization, or nation converts knowledge into capability.

The Learning Velocity Model

L — Locate

How quickly can we identify emerging threats, opportunities, technologies, and changing conditions?

Measure in days, hours, minutes and seconds.

E — Evaluate

How quickly can we determine what matters and separate signal from noise?

What is in the way of any decision?

Are we learning in the best places?

A — Align

How quickly can insights spread across teams, agencies, departments, and decision-makers?

If any insights are blocked, is this done to improve our security?

R — Respond

How quickly can knowledge be converted into decisions, actions, and measurable outcomes?

Time from moment we know to moment we impact. The full timeline is important.

N — Navigate

How quickly can we adapt when assumptions prove wrong or circumstances change?

How do we learn from failure? Are we improving our ability to anticipate?

Preparing for a World Beyond AI

AI will continue to serve as a learning accelerator. It will be joined by quantum computing, which will make it possible to simulate a problem in minutes or hours. 6G will expand how we connect billions of sensors, vehicles, drones and devices to feed AI systems. Digital twins will create virtual versions of supply chains, aircraft, power grids and more so we can test operational concepts before deployment. The physical world will become measurable in new ways, ranging from advanced sensors to satellites.

And of course, as these technologies converge, new transformational waves will be experienced.

The future competitive advantage will belong to the organization or country with the fastest and most adaptable learning system, not the most information.

Moore’s law was first proposed in 1965. Gordon Moore, co-founder of Intel, predicted that the number of transistors on a chip would double every two years, leading to major improvements in computing power. It illustrated how we compressed the cost of computation.

The skeptics said you can’t keep shrinking transistors forever or the manufacturing costs will be prohibitive or the growth rate is unsustainable.

The skeptics were wrong.

The cognitive security version of this model, which has blanks to fill in, will be something like “learning velocity doubles every X years while the cost of learning falls by half.”

It’s up to us to define this model and turn a theory into an advantage that improves security of our world.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

The Forty-Year Cyber Policy Failure Congress Refuses to Address

Late last month, the former deputy assistant director of the FBI's Cyber Division testified before the House Homeland Security Committee that the federal government should consider designating ransomware operators as terrorists and pursuing felony murder charges against attackers whose intrusions kill patients. The testimony was a serious response to a serious problem. It was also a measure of how far the cyber policy conversation has drifted from the question that would actually change the threat environment.

Terrorist designations are post-hoc. Homicide prosecutions are post-hoc. Sanctions are post-hoc. Indictments of foreign operators are post-hoc. The entire architecture of American cyber enforcement is built around consequences imposed after the harm has occurred — and for forty years, Congress has steadfastly refused to legislate the one consequence that would matter most to attackers and most to victims: the right to interrupt an attack while it is underway.

A homeowner in most American states may use deadly force to stop an intruder reaching for a television. A hospital CISO watching a confirmed exfiltration leave her network in real time may do exactly one thing: document the theft and call the FBI. If she does anything else — if she reaches one hop downstream to interrupt the transfer in progress — she has committed a federal crime under 18 U.S.C. § 1030.

This asymmetry is not the product of careful legislative deliberation. It is the product of forty years of legislative avoidance. And the avoidance, I will argue, is the most consequential cyber policy choice the United States has ever made.

A legislative record without a victim

Congress has not been idle on cyber. Since the mid-1980s, it has produced a continuous body of federal cyber legislation that is, by any reasonable measure, substantial.

The Computer Fraud and Abuse Act was enacted in 1986 and amended in 1994, 1996, 2001, and 2008. The Computer Security Act of 1987 (Public Law 100-235) established NIST's authority over federal civilian computer security and, in the process, drew the jurisdictional line between civilian and national-security systems that still governs federal cyber organization today. The Federal Information Security Management Act passed in 2002 and was modernized in 2014. The Cybersecurity Information Sharing Act was enacted in 2015. The Cybersecurity and Infrastructure Security Agency was stood up as an operational component of DHS in 2018. The Office of the National Cyber Director was established by statute in 2021.

This is a Congress that has been continuously engaged with cyber for four decades. It has legislated the boundaries of federal system security. It has criminalized unauthorized access in five separate statutory revisions. It has structured the federal-private information-sharing relationship. It has built and rebuilt the organizational architecture of national cyber defense.

In forty years, it has not once legislated whether the victim of an active exfiltration has the right to interrupt the transfer.

The Active Cyber Defense Certainty Act was introduced in 2017 by Representatives Tom Graves and Kyrsten Sinema. It was reintroduced in 2019. Neither version received a floor vote. The bill's existence proves Congress knows the question is on the table. The bill's fate proves Congress has decided to keep it there.

The shape of the asymmetry

The legal vacuum has produced an operational reality that, when stated plainly, is difficult to defend.

A ransomware operator working from a non-extradition jurisdiction faces, in practice, a probability of prosecution approaching zero. Successful prosecutions of foreign ransomware operators in 2025 numbered in the low double digits worldwide, against an industry whose estimated annual revenue exceeds one billion dollars. The victim — typically a hospital, a school district, a mid-market manufacturer, a municipal government — faces the full weight of regulatory liability, civil litigation, board accountability, and operational harm.

One side of this exchange bears nearly unlimited downside risk. The other side bears nearly none. This is not a threat environment. It is a market, and the market is functioning exactly as its incentive structure predicts.

The conventional response is to point to the things we have done. The Treasury Department has sanctioned mixers and exchanges. DOJ has clawed back ransom payments, most notably the partial Colonial Pipeline recovery. FBI and partners have disrupted Hive, LockBit (twice), and the ALPHV/BlackCat infrastructure. CISA has improved baseline guidance. None of this is nothing. All of it, taken together, is too small.

These are tactical wins inside a strategic loss. Sanctions disrupt laundering for measurable but brief windows before volume routes around them. Takedowns are followed by re-branding inside a quarter. Indictments of foreign operators function as press releases. The asymmetry between attacker risk and defender risk is not closing. It is widening.

What the "next hop" means, and what it doesn't

Let me be precise about the legal change I am arguing for, because precision is the only thing that protects this argument from being misread as a call for vigilantism.

I am not arguing for hack-back authorities. I am not arguing for retaliation. I am not arguing for the right to compromise an attacker's infrastructure as a punitive measure, to recover data through offensive operations, or to engage in any conduct whose purpose is to inflict harm on the attacker.

I am arguing for the legal recognition of a category that exists in every other domain of self-defense and exists nowhere in cyber: the right to interrupt a crime in progress.

When an exfiltration is underway, the defender can typically observe the immediate next hop — the command-and-control server, the staging system, the relay — through which the data is transiting. Current law permits the defender to log this traffic, to characterize it, to share indicators of compromise, and to report it. Current law forbids the defender from taking any action against that next-hop system to interrupt the transfer in progress, even when attribution to the attacker's infrastructure is unambiguous and even when the action contemplated is narrowly scoped to interrupting that specific transfer.

This is the gap. Not punishment. Not retaliation. Interruption.

The doctrinal analogue is the long-settled law of defense of property and defense of self. American common law has never required a victim to wait until a crime is completed before responding. The reasonableness standard — proportionality, immediacy, scope — is the mechanism by which we distinguish legitimate interruption from vigilantism. We apply this standard to homeowners, to merchants, to security guards, and to law enforcement. We have declined, uniquely, to apply it to cyber defenders.

The objections, and where they fail

The standard objections to active cyber defense are serious and I want to take them seriously.

Attribution is hard. Sometimes. It is also sometimes trivial. The exfiltration to a known command-and-control server with a known operator and a known wallet, observed in real time from the victim's own network, does not present the attribution problem that the objection imagines. The objection conflates the hardest cases with all cases. A reasonableness standard — the same standard we apply in every other domain of self-defense — would distinguish them.

Collateral damage is real. Yes. The attacker's infrastructure frequently transits compromised third-party systems — hospitals, universities, small businesses whose servers have been weaponized without their knowledge. An action against the next hop could disrupt the operations of an innocent party. This is a genuine concern. It is also a concern that applies, in different forms, to every domain of self-defense we currently permit. The legal response is not prohibition. The legal response is a proportionality requirement.

The CFAA was written for good reasons. It was. The CFAA in 1986 was a response to a specific set of harms — unauthorized access, fraud, malicious intrusion — that the existing criminal code did not adequately address. Its drafters were not contemplating the question of whether a victim observing real-time exfiltration has any right to interrupt the transfer. They could not have been. The threat environment that question arises in did not yet exist. A statute written for one purpose, applied four decades later to a question its drafters did not contemplate, is not legislative wisdom. It is legislative inertia.

Active defense will escalate. Possibly. The same argument was made against every expansion of self-defense doctrine in American legal history. The empirical question of whether a narrowly defined interruption right would produce more harm than it prevented is exactly the question Congress has declined to investigate, by declining to hold the hearings, declining to advance the bill, declining to commission the study.

What the silence costs

The forty-year silence on this question is not a neutral position. It is itself a policy choice, and the choice has a price.

The price is paid in the asymmetry. Every additional year the question goes unanswered, the gap between attacker risk and defender risk grows. The ransomware industry's revenue trajectory is not a mystery and it is not unpredictable. It is a rational market response to a legal environment in which the cost of attacking is approximately zero and the cost of defending is approximately unlimited.

The price is paid in moral coherence. A legal regime that permits deadly force in defense of a four-hundred-dollar television and forbids software-based interruption in defense of a hospital's entire patient record system is not internally consistent. The inconsistency does not become coherent because we have grown used to it.

The price is paid in deterrence. Deterrence requires consequence. There is no deterrence in cyber today, against any actor of any sophistication, because there is no consequence. The consequence that matters most — the one the attacker actually fears — is interruption of the operation in progress. Sanctions, indictments, and takedowns are post-hoc. They impose costs that the attacker can model and price in. Interruption is the consequence the attacker cannot model, because the attacker does not know when, by whom, or how it will arrive.

That is the consequence Congress has declined to authorize for forty years.

A modest proposal

I am not proposing that Congress pass the Active Cyber Defense Certainty Act as written. The 2017 and 2019 versions of that bill were imperfect, and reasonable people disagreed about specific provisions. I am proposing that Congress hold the hearing.

Forty years of avoidance is enough.

The question on the table is narrow, specific, and legally tractable. Does the victim of an active exfiltration, under a reasonableness standard, have the right to take action against the immediate next hop in the transfer chain to interrupt the transfer in progress? It is a yes-or-no question. Congress has answered every other cyber question it has been asked since 1986. It can answer this one.

I expect that when Congress finally holds that hearing, the answer will involve a tightly scoped right, a high reasonableness standard, a mandatory reporting requirement, and meaningful liability for abuse. That is what the legislative process is for. The current answer — that the question is too uncomfortable to ask — is not a legal position. It is an abdication.

The grandmother in Ohio has more enforceable rights tonight than the hospital CISO watching her patient records leave the building.

That is not a security policy. That is a forty-year-old silence.

It is time to break it.

The author is a former Commander of the U.S. Army Computer Emergency Response Team with 25 years experience in information technology, cyber operations, cybersecurity and compliance. The views expressed are his own.

A Historic Summit between China and North Korea

There was no absence of Chinese and North Korean media coverage of Chinese President Xi Jinping’s visit to North Korea for meetings with Chairman Kim Jong Un. Mr. Xi’s last visit was in 2019, and what a difference seven years makes when you are dealing with North Korea.

Mr. Xi received royal treatment during his two-day visit in June 2026. Chinese and North Korean media coverage spoke of expanding cooperation on trade, agriculture, construction, and technology and enhancing exchanges in diplomacy, law enforcement and military affairs. Interestingly, China’s Minister of Defense, Dong Jun, and his North Korean counterpart, No Kwang-Chol, also participated in some of the meetings.

During the visit to the North Korea-China Friendship Tower to honor fallen soldiers from the 1950-1953 Korean War, Messrs. Xi and Kim stressed the importance of carrying forward their traditional friendship, with the “spirit of resistance against the U.S.”

Discounting the symbolism, what did China accomplish from the visit, the first foreign trip in 2026 for Mr. Xi?

1. Reaffirmation that North Korea is China's only ally, pursuant to the 1961 Treaty of Friendship, Cooperation and Mutual Assistance. Following Mr. Xi’s visit, it’s likely North Korea will receive an impressive gift from China.

2.North Korea’s recommitment to a one-China policy over the Taiwan issue.

3.A message to Russia’s Vladimir Putin that China will ensure that North Korea remains economically and strategically tethered to China. North Korea keeps China apprised of their military support to Russia for its war in Ukraine.

4.A message to President Donald Trump that China retains considerable leverage over North Korea and any U.S. effort to seek rapprochement with North Korea will be transparent to Beijing and dependent on China’s support.

What did North Korea accomplish from the visit of Mr. Xi?

1.De facto recognition that North Korea is a nuclear weapons state.

2.The prestige accrued to Mr. Kim from the visit of Mr. Xi.

3.A likely uptick in trade and agricultural assistance, pursuant to commitments Mr. Xi made during the visit. Also, likely closer collaboration between the Korean People’s Army and China’s People’s Liberation Army.

Clearly, what Mr. Kim got from the visit was the prestige of having two great powers – China and Russia – vie for North Korea’s attention. The September 2025 Victory Day celebration in Beijing, with Mr. Kim standing next to Messrs. Xi and Putin, preceded by the visit of Mr. Putin to Pyongyang and now Mr. Xi’s visit to Pyongyang all contribute to an enhanced profile for a North Korean leader who is building more nuclear weapons that can be mated to ballistic missiles that can target South Korea, Japan, and the U.S.

It’s likely the U.S. was discussed in private discussions between these two leaders. Mr. Kim probably made the case for North Korea’s nuclear program and counseled against China overtly again calling for the denuclearization of the Korean Peninsula.

In fact, Mr. Kim probably made it clear to Mr. Xi that the construction of the new Uranium Enrichment facility at Yongbyon was a message he was sending to the U.S. : North Korea’s nuclear program will make it clear (to the U.S.) that the Yongbyon nuclear complex now has a new facility that is spinning sophisticated centrifuges to produce more Highly Enriched Uranium (HEU) for nuclear weapons. Indeed, the Yongbyon facility was the facility that Mr. Kim offered to dismantle in February 2019 at the Hanoi Summit with Mr. Trump, in exchange for the lifting of sanctions imposed in 2016. Mr. Kim’s message to the U.S.: You missed an opportunity and now we will use this new HEU facility to build more nuclear weapons.

It is interesting that Mr. Xi’s last visit to North Korea was in June 2019, a few months after Mr. Trump’s summit with Mr. Kim in Hanoi. It’s clear that China was expecting a briefing on the summit and the status of U.S. relations with North Korea.

Having a friendly North Korea as a buffer state with a South Korea aligned with the U.S. is important to Beijing. Expecting China to facilitate contact between Mr. Trump and Mr. Kim is unrealistic and not in China’s interest. The ball is in the U.S. court.

The author is a former associate director of national intelligence. All statements of fact, opinion or analysis expressed are those of the author and do not reflect the official positions or views of the U.S. government. Nothing in the contents should be construed as asserting or implying U.S. government authentication of information or endorsement of the author’s views.

This article was originally published in The Washington Times and is republished here with the author's permission.

DNI Day One: Three Strategic Decisions for National Security Evolution

Authors’ Note:

This paper is intended to frame a discussion, not settle one.

Too often, debates about intelligence reform begin with organizational charts and predetermined solutions. We believe the more important starting point is identifying the strategic decisions that should not be avoided. The next Director of National Intelligence will have to address questions surrounding enterprise leadership, resource alignment, technological modernization, and strategic competition. This first paper examines those issues and the questions that should be asked. A second paper will explore potential answers.

Introduction

The next Director of National Intelligence inherits an Intelligence Community facing simultaneous technological, geopolitical, and institutional disruption.

The post-9/11 reforms that created the DNI helped solve many of the Intelligence Community's integration challenges. Those reforms were designed for a world still shaped by the aftermath of 9/11, where the primary concern was improving information sharing and coordination among intelligence agencies. They did not anticipate the strategic environment the next DNI will face.

Today, artificial intelligence is changing and improving how intelligence is collected, analyzed, and consumed. Commercial providers increasingly own capabilities once reserved for governments. Space has become a critical intelligence domain and a contested warfighting environment. Adversaries exploit the seams between foreign and domestic authorities, using cyber operations, influence campaigns, technology theft, and economic coercion to achieve strategic objectives below the threshold of armed conflict.

At the same time, the Intelligence Community faces growing fiscal pressures associated with the costs of advanced technologies, commercial data, and modern collection systems. The Community must modernize while sustaining a full operational tempo. Intelligence professionals continue to support ongoing crises and competition across the Middle East, Russia's war against Ukraine, strategic competition with China and tensions surrounding Taiwan, challenges throughout the Southern Hemisphere, and emerging frontiers such as the Arctic and space. The next DNI will not have the luxury of choosing between today's missions and tomorrow's investments; success will require doing both.

This paper identifies three strategic decision areas that deserve immediate attention from the next DNI. The question is no longer whether the Intelligence Community must evolve. The question is whether it can evolve fast enough to stay relevant.

Scenario 1:
Resist the Urge to Build New Bureaucracies

Core Question

Is the Intelligence Community suffering from a lack of organizations—or a lack of integrated management?

Discussion

Since its creation in 2004, ODNI has steadily taken on missions, oversight responsibilities, and coordinating functions. While these additions addressed legitimate needs and emerging challenges, they also have raised a more fundamental question: has ODNI become too focused on expanding its own mission responsibilities and not focused enough on its primary responsibility to lead, integrate, and support the existing functions of the Intelligence Community?

The DNI was created to serve as the nation's senior intelligence integrator, bringing together the capabilities, expertise, and resources of a diverse Intelligence Community. Its value was never intended to come from building large operational organizations or owning missions. Its value comes from setting priorities, convening stakeholders, aligning resources, resolving disputes, and ensuring that the Intelligence Community functions as a coherent enterprise.

As new challenges emerged, the Intelligence Community often responded by creating new centers, offices, governance structures, and oversight mechanisms. While many have delivered value, the cumulative effect has been to draw ODNI toward mission execution and away from its core service imperative: enabling the success of the organizations that conduct intelligence collection, analysis, operations, and support to policymakers and warfighters every day.

The central question for the next DNI is whether ODNI is best positioned as another mission-focused organization within the Intelligence Community or whether it should recommit to its original role as the leadership, integration, and convening body for the enterprise.

The next DNI should examine whether the current structure is optimized to lead the Intelligence Community as a unified enterprise or whether a renewed focus on Community Management would better support integration.

KEY ISSUES:

·Should ODNI return to a Community Management model focused on enterprise leadership, integration, and resource alignment?

·Has the growth of ODNI strengthened the Intelligence Community or diluted ODNI's ability to serve as the enterprise integrator?

·What functions are uniquely appropriate for ODNI, and what functions should remain with existing intelligence agencies and departments?

·How should the DNI exercise leadership across the Intelligence Community without creating additional layers of bureaucracy?

·What mechanisms are needed to align collection, analysis, technology adoption, workforce development, partnerships, and budgets across the enterprise?

·Should National Intelligence Managers be empowered as true enterprise leaders responsible for integrating mission execution across agencies?

·How should ODNI lead the integration of emerging priorities such as Artificial Intelligence, commercial data, and space capabilities without becoming the owner or operator of those missions?

·What is the appropriate role of ODNI in supporting Defense Intelligence, NCTC, NCSC, and other mission organizations while avoiding duplication of effort?

Scenario 2:
Resource the Intelligence Community for the AI and Space Age

Core Question

Can the United States afford fragmented investment decisions in an era where artificial intelligence, commercial data, and space capabilities are becoming decisive intelligence advantages?

Discussion

In Washington, strategy ultimately becomes a resource question. Priorities, authorities, organizational responsibilities, and technology adoption all follow the allocation of resources. In FY2026 alone, the Administration requested approximately $115.5 billion for intelligence activities, including $81.9 billion in the National Intelligence Program and $33.6 billion in the Military Intelligence Program, increases of $8.5 billion and $5.4 billion respectively over the previous year. One of the most important issues facing the next DNI is whether the Intelligence Community's current resource structure is aligned with how intelligence is actually produced, consumed, and operationalized today.Complicating this challenge is the longstanding divide between funding in the National Intelligence Program (NIP) and the Military Intelligence Program (MIP). While often treated as distinct funding streams with separate constituencies, the reality is considerably more complex. The current arrangement is rooted less in strategic design than in a series of historical compromises intended to balance the authorities of the Secretary of Defense, the Director of National Intelligence, and the broader Intelligence Community.

Today, that framework increasingly obscures more than it clarifies. What many outside the Intelligence Community do not realize is that substantial portions of intelligence resources funded through the National Intelligence Program directly support military operations, combat support activities, and defense intelligence functions. Likewise, many intelligence capabilities developed to support military missions provide strategic warning, indications and warning, and national-level intelligence for policymakers across the government.

The distinction between "national" and "military" intelligence made more sense in an era when intelligence missions, collection platforms, customers, and operational environments were more clearly separated. Today, the same satellite constellation may support strategic warning, operational planning, targeting, humanitarian assistance, maritime awareness, and battlefield operations. Commercial data is consumed by analysts and warfighters alike. Artificial intelligence models may support national policymakers, combatant commanders, and tactical operators simultaneously.

The more important question is not how funding is labeled, but whether the Intelligence Community is organized to make coherent investment decisions across the enterprise. The next DNI should examine whether the current MIP-NIP construct encourages integrated capabilities and enterprise modernization or reinforces organizational boundaries that no longer reflect operational reality.

This question has become increasingly urgent as the Intelligence Community enters a period in which technological advantage may matter as much as traditional intelligence tradecraft. Artificial intelligence, commercial data, advanced sensors, and space-based capabilities are reshaping how intelligence is collected, analyzed, and consumed. The challenge is no longer simply acquiring information. The challenge is processing, validating, integrating, and acting upon unprecedented volumes of data at operational speed.

At the same time, the Intelligence Community faces fiscal pressures and rapidly rising costs associated with artificial intelligence infrastructure, commercial data, cloud computing, advanced collection systems, and space capabilities. The challenge is establishing common priorities, investment strategies, and architectural principles across an Intelligence Community that remains divided among multiple agencies, departments, funding streams, priorities, and acquisition authorities.

For much of the Intelligence Community's history, the government drove intelligence innovation through its own research, development, and procurement activities. Today, that dynamic has fundamentally changed. Artificial intelligence, commercial remote sensing, cloud computing, advanced analytics, and many of the data sources increasingly relied upon by intelligence professionals are being developed and scaled by industry. For the first time in Intelligence Community history, a significant portion of future intelligence advantage will be derived from capabilities developed, built, and funded outside government.

This reality presents both opportunity and risk. The opportunity lies in unprecedented access to innovation, competition, and commercial investment. The risk is fragmentation: duplicative purchases, incompatible architectures, inconsistent security standards, vendor lock, and missed opportunities to leverage enterprise buying power. At no point in Intelligence Community history has the opportunity for partnership with industry been greater. Equally, at no point has the need for enterprise discipline been more important.

Taxpayers should expect intelligence resources to produce secure, interoperable, and mission-driven capabilities regardless of which budget line funds them. Achieving that outcome will require greater alignment across intelligence priorities, acquisition decisions, technology architectures, and commercial partnerships than the current construct was originally designed to support.

Key Issues

• Would National Intelligence Program and Military Intelligence Program merger be more cost effective or is increased alignment enough?

• The Department of Defense is the largest consumer of intelligence, owns most of the nation's collection infrastructure, and operates the largest intelligence enterprise. What is the appropriate relationship between the DNI, OUSDI&S, the Joint Staff, and Combatant Commands in setting intelligence priorities?

• Who should establish enterprise collection investment priorities?

• How should the Intelligence Community approach commercial GEOINT, commercial data, and emerging commercial intelligence capabilities?

• What is the right balance between AI-enabled analysis and human expertise?

• What role should open-source and commercially available information play in future intelligence architectures?

• How can the Intelligence Community modernize technology and data architectures while avoiding duplication, vendor lock, and fragmented acquisition strategies?

• What common standards, security frameworks, and enterprise investments are required to maximize taxpayer value and mission effectiveness?

Scenario 3:
Competing Without Borders

Core Question

Are U.S. intelligence institutions aligned with the realities of modern strategic competition?

Discussion

America's principal adversaries do not recognize the traditional boundaries between foreign intelligence, domestic security, law enforcement, cyber operations, economic competition, and influence campaigns.

China conducts technology acquisition, influence operations, cyber espionage, economic coercion, and military modernization as part of a coordinated national strategy. Russia blends intelligence operations, cyber activities, disinformation, political influence, and proxy networks to shape perceptions and undermine democratic institutions. Iran and other actors increasingly exploit digital platforms, transnational networks, and non-state actors to advance strategic objectives while remaining below the threshold of armed conflict.

These activities are not isolated intelligence challenges. They are components of long-term campaigns designed to influence decision-making, shape global narratives, acquire technology, weaken alliances, and gain strategic advantage without resorting to conventional war.

Yet many of the institutions responsible for defending the United States remain organized around distinctions that are becoming increasingly difficult to separate in practice even if rooted in well-intentioned constitutional constructs. Foreign intelligence, counterintelligence, law enforcement, economic security, cyber defense, and influence operations often fall under different authorities, organizations, and policy frameworks despite being employed simultaneously by America's adversaries.

The challenge for the next DNI is not simply improving intelligence collection. It is confronting the reality that America's principal adversaries have learned to exploit the seams between intelligence, law enforcement, counterintelligence, cyber defense, economic security, technology protection, and influence operations. These gaps are no longer theoretical vulnerabilities. They have become operational opportunities for foreign adversaries seeking strategic advantage below the threshold of armed conflict.

While America's adversaries increasingly pursue integrated national campaigns, the United States often responds through separate organizations, authorities, and policy frameworks. Foreign intelligence, counterintelligence, economic security, cyber defense, influence monitoring, and law enforcement remain divided among multiple institutions, even as adversaries employ them simultaneously.

For much of the post-Cold War era, the Intelligence Community focused on warning, collection, and analysis. Those functions remain essential, but strategic competition demands something more. The Intelligence Community must help identify, expose, disrupt, and provide options to impose costs on adversary campaigns that span diplomatic, economic, informational, cyber, and military domains. This requires not only intelligence excellence, but also stronger integration across government and a renewed focus on irregular warfare and strategic competition.

Key Issues

·Are U.S. intelligence, law enforcement, and national security institutions organized to compete against adversaries that operate across traditional bureaucratic and legal boundaries?

·How should the Intelligence Community support whole-of-government campaigns for strategic competition and irregular warfare?

·What role should counterintelligence play in protecting U.S. technology, research institutions, critical infrastructure, and economic competitiveness?

·How should the United States balance civil liberties and openness while countering foreign influence, manipulation, and information operations?

·Is the current division between intelligence and law enforcement authorities optimized for the realities of modern strategic competition?

·What role should economic security and technology protection play in Intelligence Community priorities and how does ODNI handle this work?

·Should the United States consider new institutional models, including an expanded counterintelligence architecture or an MI5-like construct, to address foreign influence and hostile state activity?

Conclusion

A successful DNI will not be measured by how many organizations are created, merged, renamed, or eliminated.

Success will be measured by whether the Intelligence Community becomes more integrated, more technologically agile, more operationally relevant, can defeat our adversaries, and be more capable of providing options for decision-makers during strategic competition.

The next DNI's greatest challenge will not be managing the Intelligence Community inherited from 2004. It will be preparing the Intelligence Community required for 2045.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

AI Is Speeding Up Intelligence, But Not the System Around It

AI is compressing parts of the intelligence cycle, but modernization is occurring unevenly across collection, analysis, validation, dissemination, and policymaker integration. The resulting friction—not the technology itself—creates the defining opportunity for IC leaders.

It’s tradecraft, not technology, that is a primary constraint on intelligence performance in the AI era. This piece examines where that constraint is already being tested and what IC leaders can do about it.

Compression is Already Happening

In some areas, AI is already reshaping intelligence work in meaningful and measurable ways. Former NGA Director Vice Adm. Trey Whitworth (Ret.) has repeatedly highlighted how AI is revolutionizing GEOINT. Full-motion video analysis that once required extensive manual exploitation is increasingly automated and continuous. Project Maven fundamentally changed the economics of GEOINT warfighter support by applying computer vision to operational imagery workflows. Some AI-generated products are being disseminated to senior policymakers with minimal human involvement.

Even before Anthropic’s game-changing Mythos product, SIGINT and cyber operations similarly benefited from AI. NSA's Human Language Technology program automates speaker identification and translation across more than 90 languages—enabling analysts to triage millions of intercepted communications and focus only on the relevant fraction. Cyber Command and NSA increasingly operate in what Former NSA Director Gen. Paul Nakasone (Ret.) called "persistent engagement"—environments where collection, analysis, decision-making, and cyber effects occur continuously rather than sequentially.

Open-source intelligence has arguably made the strongest strides. During recent conflicts, policymakers leveraged commercially available satellite imagery, social media, and public telemetry data in near real time. CIA's OSIRIS platform uses LLMs to synthesize vast volumes of open-source data, deliver summaries, and support analyst engagement through a chatbot. Former Open Source Enterprise Director Randy Nixon argued that these advances enabled OSINT to become “the INT of first resort”—a model for all-source intelligence collection and analysis.

But Compression Is Uneven

GEOINT, SIGINT, and OSINT lend themselves to AI adoption: they are data-rich, measurable, and in OSINT's case, unclassified. Clandestine tradecraft and rigorous analytic tradecraft are harder to accelerate.

The Beginning of a Strategy

Deputy Director Michael Ellis recently said that CIA expects AI to become an everyday “co-worker” for analysts within the next few years. He described a future where AI systems help analysts draft reports, identify patterns across massive datasets, test conclusions, and surface threats. Ellis also said that analysts are already experimenting with how to evaluate, validate, and cite AI-enabled insights. Questions that were largely theoretical only a few years ago are becoming practical tradecraft challenges:

How should AI-assisted analysis be sourced and validated?
What level of confidence should accompany machine-generated insights?
How should analysts distinguish between AI-enabled synthesis and human judgment?
What standards should govern the use of AI-generated content in finished intelligence?

These are important developments because they signal that intelligence leaders are thinking about how technology adoption requires tradecraft modernization.

Coordination, Validation, and Analytic Workflows

Deploying AI tools for isolated analytic tasks (e.g., search, discovery, drafting) is relatively straightforward. An analyst may now receive machine-generated correlations in seconds yet still wait hours or days for cross-agency coordination, sourcing validation, or product approval. Reimagining those surrounding workflows—how information moves, how trust is established, how products are reviewed, and how analysts interact with machine-generated outputs—is substantially more difficult.

The opportunity is enormous but requires redesigning the processes themselves—while continuing to deliver on policymakers’ daily needs.

Policy Integration and Decision Support

The compression challenge becomes even more visible when intelligence intersects with policymaking.

The traditional model of intelligence dissemination was built around periodic delivery and daily briefing cycles like the President’s Daily Brief. However, policymakers now consume intelligence alongside operational updates, open-source reporting, and social media—and make decisions at the edge, often faster than traditional dissemination cycles allow. To adapt, intelligence agencies will need to provide continuously updated context, machine-assisted forecasting, and dynamic collaboration embedded directly into policymaking workflows.

Consider what this looks like in practice: a combatant commander or ambassador can query an AI-enabled analytic system for a continuously refreshed threat picture, stress-test an assumption against alternative scenarios, and receive a validated assessment—all in the 30 minutes before he walks into a meeting with his foreign counterpart. Elements of this reality exist today. During Operation Epic Fury, the 38-day air campaign against Iran, AI synthesized targeting data across the battlespace in real time to support strikes on roughly 13,000 targets in just over a month—a pace of machine-assisted decision-making with no precedent in U.S. operations.

The challenge for intelligence leaders is driving development and adoption to make it systematic, trusted, and governed. As OSIRIS proved, AI-enabled platforms are already beginning to empower policymakers to interact with intelligence this way. In that environment, intelligence is no longer a product delivered to decision-makers. It is the environment in which they decide.

That does not mean abandoning rigor or replacing strategic analysis with real-time reporting. In fact, the opposite is true. As information velocity increases, the value of trusted analytic judgment, validation, and expert perspectives will increase.

The Emerging Risk: Asynchronous Modernization

The danger is that different parts of the system are modernizing at different speeds. Accelerating functions does not eliminate friction between functions. In some cases, it can increase it.

·Faster collection can overwhelm coordination processes.

·Faster analysis can outpace dissemination workflows.

·Automated insight generation can challenge validation, trust, and decision integration.

The Leadership Challenge

Leading companies discovered that AI could not remain a standalone innovation initiative. As AI began reshaping workflows, governance, and strategy, responsibility migrated from CIOs and innovation teams to CEOs and boards. Intelligence leaders must make a similar pivot. Modernization cannot be outsourced to technologists, innovation offices, or isolated teams. Pockets of ‘AI money’ and lists of ‘AI projects’ are technocratic, not strategic.

To fundamentally modernize how the system operates, IC leaders must be directly involved in reimagining the intelligence cycle, redefining tradecraft expectations, reshaping decision models, and aligning institutional incentives. That means three things.

First, leaders must connect the technology agenda to the mission agenda. The private sector learned that AI transformation fails when it is treated as an IT problem rather than a strategic one. The IC faces the same risk. The goal is decision advantage at the speed of policymakers and warfighters. Keeping that mission orientation at the center of the technology agenda is a leadership responsibility that cannot be delegated.

Second, leaders must own the coherence problem. It is not enough to authorize AI investments and track deployment metrics. Leaders must make architectural choices about which parts of the cycle to accelerate together, how to manage the seams between them, and what governance structures are needed to ensure the system absorbs new capability without creating new failure modes.

Third, leaders must redefine what tradecraft means in an AI-assisted environment. The standards that govern sourcing, analytic confidence, and product integrity were built for human workflows. They need to be deliberately redesigned—not abandoned—for an environment where machine-generated insights are embedded throughout the production chain. Deputy Director Ellis’s four questions are the right starting point.

AI may compress intelligence production, but only leadership can compress the distance between insight and decision.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Why Human Intelligence Matters More in an AI World

An impending casualty of artificial intelligence, we are told, is the human spy. The conventional wisdom is that in our AI future, there’s little need to recruit agents, plan secret rendezvous, or conduct dead drops—the old-school tradecraft of espionage. “Human spies in the field will become rare,” wrote David Ignatius after surveying the growing field of AI intelligence startups. “The future of espionage is written in zeros and ones.”

Yes, AI will become an indispensable tool for case officers, agents, fabricators, counterintelligence services, and the rest of the intelligence world. But this will have the paradoxical effect of increasing the importance of old-fashioned human intelligence.

Economics tells us that the value of something is determined by the benefit it provides on the margin—not by its raw power or pervasiveness. We should expect that AI will lower many barriers to technical intelligence collection and analysis. This has happened in other spheres: Open-source encryption put tools once reserved for states into everyone’s hands. Likewise, you no longer need a national space program — just a credit card — to order up high-resolution satellite photography and buy AI-powered software to analyze the results.

Human intelligence is scarcer and harder to replicate, and so its marginal value will rise. Vast amounts of data alone cannot reveal what a foreign leader intends to do, nor what is happening inside an “air-gapped” network. Human sources can.

Human intelligence has always helped validate technical collection. That role will grow in importance as AI poisons the information environment. Electronic channels will be flooded with fabricated phone calls, forged documents, and other convincing synthetic media. Signals intercepts will be harder to trust on their own. A human agent can help cut through that deluge of disinformation—and help distinguish authentic source material from forgeries.

Finally, AI is making digital communications less trustworthy, as we already see with scammers using deepfake images, audio, and video calls for fraud. As deepfakes get harder to detect, the rational response will be to distrust electronic communications. The corollary? Secure, person-to-person communication will rise in value. Case officers have long relied on dead drops, brush passes, and brief meetings to communicate with their agents. In an AI-saturated world, these traditional techniques may be the most reliable methods of communication we have.

None of this means that technical intelligence will become unimportant. And AI will certainly transform human intelligence operations—for better and for worse. It will help case officers select operational sites, prepare for agent meetings, and improve their ability to persuade foreigners to commit espionage. At the same time, AI will supercharge surveillance, provide fabricators with an unlimited source of plausible but false information, and help counterintelligence services predict the activities of case officers operating in their countries.

Still, as AI makes technical collection cheaper, deception easier, and digital communication less trustworthy, the value of human sources will rise. And the case officer, working in the shadows of our artificial future, will matter more than ever.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

The AI Race Won't Be Won by the Best Model—But by the Fastest Military

The United States Intelligence Community does not ordinarily deal in hyperbole. When the Office of the Director of National Intelligence released its annual threat assessment on March 18, artificial intelligence featured prominently among a complex array of challenges — elevated, alongside quantum computing, to what the IC now treats as a central driver of power and strategic risk.

The document concluded that Beijing “is the most capable competitor in this field” and aims to displace the United States as the global AI leader by 2030, warning that AI has already been used in recent conflicts to influence targeting and streamline decision-making.

How Washington responds, and whether its current posture matches the urgency its own intelligence community has outlined, came into sharp relief in May, as President Trump touched down in Beijing alongside Nvidia CEO Jensen Huang for a high-stakes summit that put AI and semiconductors at the center of United States-China diplomacy.

The three-month problem

According to some assessments, open-source Chinese AI models currently trail the most advanced American proprietary systems by roughly three to six months — DeepSeek’s own V4 release acknowledged that gap, which broader estimates from Epoch AI place closer to seven months. Yet that sliver of a lead is effectively meaningless from a military standpoint, because it takes not months but years for armed forces to absorb AI technology and translate it into a genuine battlefield advantage.

Carlos Perez, Director of Security Intelligence at TrustedSec, says the public numbers may not tell the full story.

“It is also important to recognize that China operates models that are not publicly available, so we have limited visibility into their true capabilities,” he tells The Cipher Brief. “Companies such as Alibaba and Tencent operate under significant government oversight and legal obligations tied to state investment and regulation. As a result, the actual capability gap may be smaller or larger than public comparisons suggest.”

China is not waiting for Washington to sort out its strategy.

The 2026 threat assessment describes Beijing’s approach as a coordinated, national-level strategy aimed at displacing the United States as the most influential AI power by 2030. Beijing is deploying autonomous drone programs at speed, integrating swarm intelligence into military doctrine, and leveraging a centralized governance architecture that allows civilian AI firms to be folded directly into People’s Liberation Army modernization efforts.

Alibaba and Baidu were both added to the Pentagon’s Chinese Military Companies list in February. DeepSeek has since become part of the People’s Liberation Army’s efforts to modernize its military healthcare infrastructure, according to analysis by the Foundation for Defense of Democracies. The House Select Committee on China concluded last year that the company was built, at least in part, on restricted American semiconductor chips and that its app functions as a direct pipeline for foreign intelligence collection on American users.

Aaron Estes, vice president of product management at Binary Defense and a former defense and intelligence official with 25 years of experience, tells The Cipher Brief the threat is more nuanced than a simple capability gap.

“The danger is that once these models are ‘good enough,’ the advantage shifts from model quality to speed of deployment, access to operational data, integration with command systems, and willingness to use AI in real-world workflows,” he explains. “A three-month gap in frontier model performance can disappear quickly if the other side is better at turning AI into operational tempo.”

Indeed, the real gap is not in the models. China can put its tech sector to work for its military tomorrow. The United States has to pass a bill first.

A policy vacuum in the middle of a race

On January 9, Defense Secretary Pete Hegseth signed the Department of War’s AI Acceleration Strategy — a document that declared speed wins, that the risks of moving too slowly now outweigh the risks of moving imperfectly, and that 2026 was the year the Pentagon would get serious about military AI dominance. Overlapping innovation offices were folded into a leaner CTO Action Group.

The document landed with fanfare. Yet, the budget did not follow. Washington keeps declaring AI a national security imperative while trimming the agencies and funding lines that would actually make it one — a contradiction the Bloomsbury Intelligence and Security Institute flagged in its recent read of the annual threat assessment.

Congress passed a more than 3,000-page National Defense Authorization Act that touched on AI across dozens of provisions — banning both the Defense Department and intelligence agencies from using DeepSeek and directing the Pentagon’s chief digital officer to build a department-wide AI assessment and procurement framework — yet defense analysts say it falls well short of the enforceable legal architecture the military needs.

Matthew Wein, a former policy advisor to the Department of Homeland Security, underscores that the White House’s March National Policy Framework for AI does not fill that gap.

“A strategy would be helpful to indicate where the government thinks we have an edge over competitors and how to maintain national security priorities as the technology landscape continues changing,” he tells The Cipher Brief. “It would also help the private sector understand how the government assesses China’s capabilities and give a blueprint for the frontier labs to use as a foundation to maintain their edge going forward.”

However, Leah Siskind, Director of Impact and AI Research Fellow at the Foundation for Defense of Democracies, argues that the threat itself is being mis-framed.

“There isn’t one AI threat, there’s a whole portfolio,” she tells The Cipher Brief. “Adversaries using AI for influence operations and cyber. Adversaries stealing or distilling American models. Adversaries racing us to military applications. A coherent strategy has to work all three at once, and right now Congress is moving faster than the executive branch on the first two.”

Siskind points to two bipartisan bills sitting on the floor — the AI OVERWATCH Act, which would codify congressional review of advanced chip sales to adversaries, and the Deterring American AI Model Theft Act, which creates a statutory framework to address Chinese model distillation attacks.

“The question is whether the White House, Commerce, and the AI czar’s office get behind them or fight them,” she continues.

The dispute between the Pentagon and Anthropic made that governance vacuum concrete. Anthropic was the first AI company to deploy frontier models on classified government networks, a position formalized by a $200 million DoD contract in July 2025. Talks broke down after the company refused to waive restrictions on autonomous weapons and mass domestic surveillance.

The Pentagon designated Anthropic a supply-chain risk—the first time that authority, historically reserved for foreign adversaries, had been applied to an American company —and Anthropic subsequently sued. As Siskind puts it: “Treating American frontier labs like Anthropic as national security supply chain risks is a strategic gift to Beijing.”

From experimentation to the edge

There is a basic problem that receives insufficient attention. Most military AI systems only work when connected to the internet. In a war, an enemy will cut that connection fast. OpenAI’s Pentagon contract actually reflects this limitation — its models can only be accessed through the cloud, meaning they cannot be built into weapons, sensors, or equipment troops use in the field.

The Army’s Project ARIA, Army Rapid Implementation of Artificial Intelligence, announced in March, explicitly aims to develop AI capabilities designed to function in denied, communications-degraded environments, but that effort remains in early development. Perez points out that AI is already deployed operationally on the intelligence and planning side, largely through platforms developed by companies such as Palantir, but stresses that the greater challenge is at the tactical level.

“Ukraine has demonstrated how AI can be integrated into autonomous drone operations and real-time battlefield adaptation, and this is an area where the U.S. is still learning and evolving rapidly,” he says.

Meanwhile, the Stanford AI Index, an annual comprehensive measurement of global AI progress published in April by Stanford University’s Institute for Human-Centered Artificial Intelligence, found that the United States ranks 24th globally in AI adoption at 28 percent. Singapore sits at 61. The UAE is at 54. Those figures reflect the civilian economy, not the military, but the pattern maps uncomfortably onto the Pentagon’s own trajectory.

Estes emphasizes that fixing it requires more than a strategy document.

“The Pentagon’s biggest obstacle is not a lack of AI technology. It is the gap between prototype and trusted operational use,” he continues. “Battlefield AI needs clean data, secure deployment environments, clear authorities, auditability, human accountability, and integration into existing systems that were never designed for this pace. That is where most AI programs get stuck.”

The Beijing question

Against that backdrop, President Trump’s arrival in Beijing in May was particularly important. He landed with a delegation that included Apple CEO Tim Cook, Tesla CEO Elon Musk, and Nvidia CEO Jensen Huang — the latter a last-minute addition whose presence signaled that semiconductors and AI were central to the agenda alongside tariffs, rare earths, Taiwan, and the Iran war.

Treasury Secretary Scott Bessent announced that Washington and Beijing would establish a protocol on AI best practices aimed specifically at preventing the most powerful models from falling into the hands of non-state actors. Pressed on why substantive dialogue with China was even possible, Bessent was candid.

“The reason we are able to have wholesome discussions with the Chinese on AI is because we are in the lead,” he responded. “I do not think we would be having the same discussions if they were this far ahead of us.”

How durable that lead is remains an open question. The Commerce Department has cleared roughly 10 Chinese companies — including Alibaba, Tencent, ByteDance, and JD.com — to purchase Nvidia’s H200 chips. However, no deliveries have been made: Beijing has quietly been steering its major tech platforms away from purchases to support domestic chipmakers instead.

China hawks on the Hill see the chip approvals as proof of exactly the problem they have been warning about — that Washington has no coherent line between competing commercially with Beijing and containing it militarily. The United States named AI the defining challenge of the era. It is now debating whether to sell China the chips that define it.

Wein says the path forward requires the government to be a better customer, not just a louder regulator.

“A more lax regulatory environment may be helpful for labs’ commercial growth, but a government strategy would also give them a useful framework from one of their largest customers, and position the United States as a standard setter in the space,” he asserts.

The cost of getting that wrong is not theoretical.

“The risk of relying too heavily on general-purpose commercial AI is that we end up with tools optimized for broad productivity, not contested military environments,” Estes adds. “Military AI has to work with incomplete data, deception, adversarial manipulation, degraded communications, classified context, and life-or-death accountability. What the United States actually needs are AI systems built from the ground up for the realities of warfare, not adapted from products designed for the consumer market.”

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Can the Pentagon Buy Faster Before the Next War Arrives?

“The reasons why DoD (the Defense Department) accepts flawed business cases are both structural and cultural in nature. Poor acquisition decisions are compounded by a budget planning process that requires DoD to secure long-range funding commitments before a program’s business case is fully understood. The current process incentivizes ‘starting fast’ -- awarding massive development contracts quickly, often in the name of preserving the industrial base, and obligating funds rapidly to ensure the budget is not ‘lost’ to another program. Success is all too often measured by activity (money spent), not by outcomes (capability delivered).”

That’s a quote from a Government Accountability Office (GAO) report released June 9, entitled, Weapon Systems Acquisition: Beyond Business as Usual -- Using Leading Practices to Curb Waste and Save Billions. It describes long-term problems within the military procurement systems that have emerged from U.S. arms use in the Iran and Ukraine wars that in turn has drawn the attention of the White House, the Pentagon and the Congress.

The June 9, GAO report said, “DoD plans to invest over $2.4 trillion to develop and acquire its costliest weapon systems. The need for smart spending and increased urgency and innovation for these acquisitions are national imperatives to help DoD maintain a competitive edge over adversaries. But DoD continues to struggle with delivering timely, cost-effective solutions to the war-fighter, and slow, linear development approaches persist.”

The result, according to the GAO, is “the expected time frame for major programs to deliver an initial [operational] capability now exceeds 12 years. Every month of delay in a weapon system acquisition program causes a war-fighter to rely on aging, less-capable equipment for longer.”

The GAO found, “In contrast, leading commercial companies iteratively develop business cases to respond to users’ needs and finish fast, helping them stay on budget. They re-assess business cases regularly to avert problems sooner. They also ramp up investments as products demonstrate progress.”

In contrast, the GAO said, “DoD has yet to fully adopt these leading practices because acquisition policies do not treat iterative development as a founding principle for all weapon system acquisitions programs.”

However, as the GAO points out, “As noted in its November 2025 policy memorandum aimed at revamping the defense acquisition system, DoD now plans to maximize acquisition flexibility, among other changes.”

Perhaps the poster child for DoD procurement issues is the built F-35 aircraft, whose prime contractor has been Lockheed Martin.

The F-35 Lightning II is a family of fifth-generation strike fighters – with versions in the Air Force, Navy and Marines -- that integrate low-observable (stealth) technology with advanced sensors and computer networking capabilities. According to the GAO, DoD completed the final phase of the original F-35 development program in March 2024 -- over a decade later and at a cost of $250 billion more than originally planned. Now, DoD is upgrading F-35 capabilities by adding technology innovations under modernization efforts.

The GAO said, “In recent years, the program paid contractors hundreds of millions of dollars in incentive fees that were intended to improve on-time delivery. However, the structure of on-time delivery incentives allowed the contractor to deliver aircraft up to 60 days late, and still earn some of the fee.”

In addition, since 2017 there have been disputes over software with Lockheed claiming certain portions of the software were developed at the company’s expense and could not be used by the military unless it paid licensing fees. At the same time the Pentagon claimed that the government had paid for the F-35’s software development, and the military should not pay twice for access to it.

According to a 2025 GAO report, the Pentagon did not have the data rights needed to perform software diagnostics and some maintenance activities on the F-35s, including engine repairs, without an authorized contractor, because Lockheed Martin owned key intellectual property.

The newest GAO report on the F-35, sent last Wednesday to the House and Senate Armed Services Committees, said, “DoD operates and sustains over 800 F-35s and plans to buy about 1,700 more aircraft by the mid-2040s. DoD uses the F-35 to perform a wide range of missions and it is vital to the success of U.S. combat operations and homeland defense, according to DoD.”

GAO also said sustainment costs for the planned life of the overall F-35 program through 2088 have been estimated at $1.6 trillion.

However, the main finding of last week’s GAO report is: “Since 2021, F-35 sustainment costs have increased as fleet size has grown, but the F-35 fleet continues to not meet sustainment performance goals, with mission capable (MC) rates and full mission capable (FMC) rates declining.”

For example, according to the GAO, while operating costs have grown, the Air Force full mission capable rate for its F-35A aircraft in 2025 was 28.5 percent; the Navy F-35Cs were 15.3 percent, and the Marine F-35B and C versions were 16.2 percent and 22 percent respectively.

The GAO report said the F-35 Joint Program Office (JPO) “identified depot repair speed and the ability to procure parts as root causes driving low sustainment performance rate.” It also added that “an investment of about $13.7 billion in addition to previously planned spending through fiscal year 2031 to meet performance objectives. Overall, the strategy aims to achieve a fleet-wide [F-35] readiness level of an 80 percent MC rate and a 65 percent FMC rate by 2030.”

In 2025, GAO found that, according to JPO officials and maintainers, “the F-35 has faced significant corrosion issues that maintainers cannot repair without contractor support due to a lack of technical data.” The GAO recommended the DoD determine whether it had access to technical data needed from manufacturers, but last week’s new GAO report said they had not.

“We have consistently found,” the GAO said, “that technical data issues have hindered F-35 sustainment. While the [Defense] Department has taken some incremental steps to address these issues, significant challenges remain without a clear timeline for resolution.”

This is a problem recognized on Capitol Hill. The Senate Armed Services Committee, in its Friday press release describing the panel’s approved version of the Fiscal Year 2027 National Defense Authorization Act listed “Shifts burden of proof to contractors to justify restrictions on technical data” as one of its several major legislative reforms for the Pentagon.

Another DoD reform approved by the Senate committee: “Requires acceleration of the adoption and purchase of low-cost munitions, including a pilot program on air defense interceptors.”

Picking up on this issue last Friday was former-Joint Chiefs Chairman, Air Force General C.Q. Brown Jr. Speaking at the Center for a New American Security’s 2026 National Security Conference, Brown said, “The challenge we have is when you look at munitions, we don't prioritize those as much and they become the numbers [that] will go up and down. The numbers are enough just to keep the stockpile, but not enough to actually execute [if you are in a war].”

Brown went on to say, “Resourcing is an issue. If you can't get a budget on time, you can't write a contract.” That’s a reminder that the last time Congress passed a defense appropriations bill before October 1, when the federal fiscal year begins, was in 2010, for the fiscal year 2011 Pentagon budget. In each of the past 16 years, the DoD has had to rely on continuing resolutions and omnibus reconciliation packages, which delay defense funding for months into the fiscal year.

Brown called for “consistency and delivery” by which he meant how quickly you're able to scale numbers of items and deliver them, knowing funds are available.

“One of the things that I was pushing for as the Air Force chief and again as a [Joint Chiefs] Chairman is multi-year procurement,” Brown said.

He went on, “If you did multi-year procurement with a floor, you're always producing X number. You can always go above that, but never go below a certain number. That keeps industry with some level of certainty and that builds that trust so they can continue to deliver.”

Brown added, “How well do we understand across all our various weapon systems? Who are the primes [prime contractors] to do this work? Who are the subs [sub-contractors] and what [does]] the supply chain looks like. And then on top of that, what is the workforce or whatever automation you're going to use. It was a work I was doing as the [Joint Chiefs] Chairman to identify as we looked at all of our operational plans.”

Brown then continued, “So if you buy affordable mass [of weapons and munitions] today and put it on the shelf, will it still be viable five years from now, or five weeks from now? And so the key part here is how can you actually have a level of affordable mass that's adaptable that you can continue to upgrade over time and change out -- either software or parts or hardware and then you can scale it very quickly.”

Brown said at one point, “The aspect of [defense] production is also part of deterrence,” adding, “and showing that you can surge capability” is an important part.

As Brown and others have pointed out, Ukraine and Iran have suddenly focused more attention on innovation at the lower weaponry and munitions levels, at the same time that costly changes are continuing to be needed in the more complicated air and space armament such as the F-35.

Like it or not, the American people may have to learn to live with spending $1 trillion-plus, hopefully for a Department of Defense, rather than a Department of War.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

The Warning Paradox: Why Correct Intelligence Often Fails

In the months before Russia’s February 2022 invasion of Ukraine, the U.S. Intelligence Community did something remarkable. It got the call right—and said so out loud, rather than only in classified internal documents.

Intelligence assessments identified the likely invasion, the timing window, and plausible operational scenarios with a level of confidence the IC rarely displays outside classified channels. The CIA director flew to Moscow to deliver the warning directly. Intelligence was declassified at a pace and volume rarely seen in modern American practice. By any fair measure, the warning was a success.

Yet the policy response did not match the clarity or urgency of the warning.

That is the Warning Paradox: the recurring pattern in which intelligence correctly anticipates strategic events, but institutions fail to convert warning into preventive action before the forecast becomes a crisis.

My experience during the run-up to the Ukraine invasion showed that the Warning Paradox is not inevitable. The corporate intelligence team I led at the time reached the same conclusion. My team’s pre-invasion assessment correctly identified the likely move, the timing window, and the implications for our people, facilities, and supply chains across the region. The Wall Street Journal later documented that work on the front page of its December 12, 2022, print edition. Our assessment placed the start of hostilities within the twenty-four-hour window in which the invasion ultimately began.

Decision-makers took the warning seriously. Contingency posture was raised. People were moved. By the time Russian tanks crossed the Ukrainian border, the enterprise was in a defensible position. On the corporate side, our warnings translated into action.

The policy response in governments around the world, however, did not match the clarity or urgency of the warning.

Although sanctions planning was underway before the invasion, the most consequential measures were imposed only after Russian forces crossed the border. Lethal aid to Kyiv increased substantially only after the invasion became reality, not while it remained a contested forecast. Alliance posture stiffened in response to the war rather than in anticipation of it. The warnings were correct and early. The response was neither.

Ukraine was not an anomaly. It was one of the clearest recent examples of the Warning Paradox: intelligence got the threat right, but institutions failed to act with the urgency of the warning.

The August 6, 2001, Presidential Daily Brief did not produce a different September 11th outcome. The accumulating signal on Hamas before October 7, 2023, did not produce a different response inside the Israeli system. Crimea in 2014, COVID-19 in late-2019/early 2020, and the Afghan collapse in 2021 are each, in their own way, variations on the same theme. These cases differ in scale, context, and institutional setting. However, they point to the same recurring failure that warnings can be analytically valid and still die before they become a decision. A correct warning is not a sufficient condition for preventive action.

The bottleneck is rarely analytical. Intelligence producers often identify emerging threats with more accuracy than later public narratives allow. The gap persists between what the intelligence community sees and what the policy customer can absorb, decide on, and act on under uncertainty. That gap is structural. It persists for three reasons that appear to be features of the system, not bugs.

First is the asymmetric cost of acting on a warning that does not materialize. A policymaker who acts on a high-confidence warning that turns out to be wrong pays a visible, attributable price, often in squandered political capital, damaged relationships, or the “cried wolf” label. A policymaker who fails to act on a warning that turns out to be right pays a price too, but it is diffused across the system, shared with predecessors and peers, and absorbed by the institution rather than the individual. The incentive structure does not reward preventive boldness. It rewards explainable caution. Until that asymmetry is acknowledged and consciously offset, intelligence will continue to land in front of consumers whose private calculus rewards waiting over action.

Second is the consumer literacy gap: the uneven ability of decision-makers to understand probabilistic warnings and convert them into action. Intelligence is a product designed for a customer. Yet the customer is often a decision-maker whose training in absorbing probabilistic warnings is uneven at best. “We assess with high confidence” is a precise analytic statement; the structures to translate it into a specific decision tree, a sequenced set of preventive moves, and a named owner are not consistently present on the consumer side. This is less a criticism of individual policymakers than of the system around them, one that has invested heavily in producing intelligence and very lightly in building the institutional muscle memory to both receive and act upon it. The same pattern repeats in the corporate world. Boards are briefed on threats they could not, if asked, translate into a Monday morning action item.

The third is the governance gap between warning and decision. Inside both government and large corporate enterprises, no single office holds the authority to convert warnings into preventive action. The intelligence producer’s job ends with the warning. The policymaker’s job begins with a decision. The implementer’s job begins at execution. Between warning and decision is a coordination space nobody owns by name. This is the same governance gap that organizational risk practitioners describe in different vocabulary, and it is the most consistent reason correct intelligence fails to produce appropriate preventive action.

After nearly three decades in the Intelligence Community, including fourteen years at the Defense Intelligence Agency, I have seen what correct warning looks like when it lands well — and when it reaches decision-makers who cannot or will not convert it into action. The years since leaving public service, now leading corporate intelligence at a Fortune 50 enterprise, have taught me that the pattern is not peculiar to government. The exact vocabulary may differ, but the grammar remains the same.

The intelligence does not need to change as much as the consumer system around it does.

First, assign ownership. Someone in the room, by title, must be responsible for converting warnings into preventive action and be measured on it. In the absence of named ownership, a warning is a briefing, not a decision input.

Second, build the institutional muscle memory to receive information as a decision input. Decision-makers should rehearse the conversion of warnings into action, as operators or emergency services personnel do with contingency responses. Tabletop exercises, decision drills, and scenario planning before a crisis are inexpensive relative to the cost of surprise. They are also irregular at best in most organizations.

Third, lower the political cost of acting on a warning that doesn’t materialize. This is the hardest of the three because it is cultural rather than procedural, and it requires senior leaders, whether in government or in the C-suite, to defend subordinates who acted prudently on sound intelligence even when the worst-case scenario does not occur. Until acting on a warning is professionally survivable when the warning proves overcautious, the incentive structure will continue to punish exactly the behavior we say we want.

Chuck Randolph, a colleague in the corporate security community, puts the point more directly: “Intelligence without action is just information.” The line is uncomfortable because it is correct.

In many of the most consequential cases, the U.S. Intelligence Community is not failing simply because it cannot see the risk. The failure lies in the consumer system, which cannot act on what it is shown. Until that changes, we will continue to be surprised by events we warned ourselves were coming. We will continue to mistake analytical success for policy success, and policy failure for intelligence failure.

The Warning Paradox is not a forecasting problem. It is a consumer problem. Like most consumer problems, the people best positioned to solve it are the ones least accustomed to thinking of themselves as the bottleneck.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

America's Veterans Are a Defense-Tech Asset — We're Wasting Them

I spent my formative adult years in service to my country—from seventeen to twenty-nine—and the core of how I identify myself remains that of a United States Marine. When I separated, the playbook was clear: suit up for the boardroom. Investment banks and consulting firms were the promised land, where top talent was expected to go.

It was not an easy transition. Few banks at the time viewed a Marine infantry officer's background as preparation for a career in finance. Frustrated by what my peers and I experienced, I helped found Veterans on Wall Street in 2009—a consortium of Bank of America, Citigroup, Credit Suisse, Deutsche Bank, and Goldman Sachs, led by veterans at each institution and focused on veteran hiring, transition support, and charitable giving. It was the right response for that moment.

That moment has passed.

The world has shifted decisively beneath our feet. Artificial intelligence, autonomous systems, cybersecurity, space, and advanced manufacturing are reordering the global balance of power and creating one of the most acute talent shortfalls in modern economic history. America's strategic competition with near-peer adversaries is no longer primarily a contest of battalions. It is a race to build, deploy, and operate AI systems, autonomous platforms, and secure communications infrastructure faster and more effectively than any rival. The conflicts unfolding today across the Middle East and beyond are shaped as much by autonomous systems, electronic warfare, sensors, and AI-enabled targeting as by boots on the ground.

The organizations building these systems need people who understand not only the technology, but the operational environments in which it will be used. Veterans who have worked in signals intelligence, operated in contested communications environments, or commanded logistics chains in austere conditions bring something no computer science curriculum can replicate: they have been the end user. They understand which failure modes matter.

The numbers are striking. More than 200,000 servicemembers separate from active duty every year. Meanwhile, technology occupations are projected to grow roughly twice as fast as overall employment over the next decade, with particularly acute demand in AI, data science, cloud infrastructure, and information security—roles that remain structurally undersupplied.

Yet the existing Transition Assistance Program (TAP) often functions as a checklist rather than a tailored pathway. Only about 52 percent of servicemembers complete the recommended one-year TAP timeline—a program designed for an economy that has itself moved on from the jobs it was built to funnel veterans into.

The irony is that veterans may be better positioned for the defense-tech economy than almost any other talent cohort—if we invest in the translation layer. Their skill sets naturally lend themselves to roles where human judgment, leadership under uncertainty, and adversarial thinking are most valuable: precisely the roles least susceptible to AI disruption and most critical to national security. They did not just study modern conflict. They fought it.

This is not a resume-translation problem. It is a strategic investment problem.

I learned this in a different context. In October 2004, my battalion deployed in support of what would become Operation Phantom Fury—the Second Battle of Fallujah, the bloodiest battle of the Iraq War. Eight days before the assault, we received an attached Iraqi Army company: 36 men out of an original 146, the rest having deserted. I didn't speak Arabic. My interpreter was a 55-year-old former physics teacher. Their weapons handling was dangerous. Their loyalties were uncertain. The decision before me was whether to lead them into combat or tuck them behind our movement through the city, as my peers planned to do. I led them from the front. And what emerged was an effective fighting unit—clearing houses alongside us, gathering intelligence no one else could access, saving lives on both sides.

The lesson I carried forward: the hardest leadership decisions are rarely about resources or capability. They are about the will to build the bridge between what you have and what the mission demands. That is exactly where we stand today on veteran transition.

The emerging defense-tech sector is already recognizing this. Firms such as Anduril and Shield AI—both co-founded by veterans—are hiring aggressively from military ranks. Organizations like MVA Foundation (MilVet Angels) have backed these up and coming defense tech startups, along with others like Hermeus, Ursa Major, and Cowboy Space Corp—a portfolio that maps almost precisely onto the Pentagon's own list of mission-critical technology priorities. What makes their model distinctive is its structure: all carried interest from exits flows back into a foundation funding veteran transition and entrepreneurship programs, creating a self-reinforcing cycle between investment returns and the next generation of veteran technologists. Palantir has taken a complementary approach through its American Tech Fellowship—a high-intensity program recruiting transitioning veterans and enlisted leaders, requiring no tech degree, and connecting graduates directly with defense tech employers. The core insight mirrors the best veteran transition efforts: the most persistent barrier is translation, and veterans routinely underestimate how directly their military experience maps to the roles AI-era employers can't fill. These are promising efforts. They are not yet at scale.

What would scale look like? Three things.

First, rebuild TAP around defense technology pipelines. The NSCAI and CNAS have argued for years that the program underdelivers, but they frame it as a workforce-quality problem. It is a national security routing problem. Statute already requires counseling to begin twelve months before separation, yet GAO finds 70 percent of servicemembers miss that threshold and commanders routinely waive attendance against their own services' rules. The answer is not another reform package layered onto a $500 million interagency program. The Secretary of War should use existing authority and appropriated dollars to redirect counseling toward critical-technology tracks, strip waiver discretion below the general-officer level, and replace completion rates with twelve- and twenty-four-month placement metrics tied to commanders' evaluations.

Second, fix SkillBridge's throttling problem. More than 25,000 servicemembers participated in fiscal year 2025 across 6,000-plus partners, but GAO's 2024 review found commanders deny or discourage participation because losing someone for 180 days reads as a readiness hit while the national security benefit accrues elsewhere. The Department of War should change how participants are counted against unit manning in their final 180 days, set a service-wide floor on approval rates with denials reviewable above the immediate commander, and require outcome reporting tied to placement in critical-technology sectors. The talent bench exists; the accounting rules are what keep commanders from releasing it.

Third, the private sector must signal that this is a strategic priority, not a corporate social responsibility initiative. Emerging defense tech companies competing for government contracts should be first movers. They have both the operational need and the patriotic case. When I helped found Veterans on Wall Street, the animating insight was that the private sector had to lead—that institutions with resources and relationships had to build the bridge before the government could cross it. The same logic applies now, in a sector with far higher national stakes.

A generation ago, top talent was expected to go into finance and consulting. Today, technology—and America's defense-tech companies—are building the arsenal of democracy for the 21st century. The people who know best how that arsenal must perform are already among us: 200,000 men and women separating from service every year, looking for someone to show them where they fit in the new economy.

The answer is in front of us. We just need the will to build the pipeline.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

Economic Security in an Age of Strategic Competition

There is a growing perception among long-standing US allies that they need to expand commercial relations with the People’s Republic of China (PRC). A thaw or détente with the PRC brings both rewards, particularly for a sluggish economy like Britain’s, but also major risks. History shows that a superpower can ruthlessly exploit détente with the West.

Economic Security and Intelligence

Economic security and intelligence are nothing new. Before the Second World War, for example, Britain ran a small outfit known as the Industrial Intelligence Centre (IIC). Run by a former MI6 officer, Desmond Morton, the IIC provided a coordination of intelligence on German rearmament and, working with MI5, assessed Britain’s commercial vulnerabilities. British intelligence helped to devise the UK government’s War Book, which set out emergency regulations to protect critical national infrastructure in the event of war. After the Second World War, during the Cold War, it became a staple of British and other western intelligence to assess the size and strength of economies behind the Iron Curtain.

Risky Business

In 1972 President Nixon and his national security advisor, Henry Kissinger, ushered in a policy of détente with the Soviet Union. Its purpose was to further divide the Soviet Union from China. Papers at the Nixon library show that Kissinger was under pressure from British firms, in particular, to open up markets behind the Iron Curtain. Britain was in a dire economic doldrum following an oil shock due to a war in the Middle East.

Kissinger and Nixon knew that not all commercial technologies could be transferred to America’s main strategic enemy, the Soviet Union. The White House was accurately afraid of dual use technologies, namely those that were civilian but could also be used for military purposes. Kissinger limited the sale of high end computers and microchips, for example, only allowing second tier components to be sold to the Soviets.

Although Nixon and Kissinger accurately guessed that US industries would be targets for Soviet espionage, the extent to which the Soviets exploited détente would have been beyond their wildest imaginations. The collection of scientific and technical intelligence from the US was conducted by Soviet military intelligence (GRU) and the KGB, whose operating arm, Line-X, reported to Directorate T (Technology). In 1973 the KGB assigned an officer to New York whose full-time job was to collect (steal) US scientific and technical intelligence (S&T). By 1980 the US was producing more S&T intelligence for Moscow than the rest of the world combined. Visiting Soviet trade delegations to US research centers, laboratories and fortune 500 companies, for example, were packed with undeclared Soviet intelligence officers. In an agricultural delegation of a hundred Soviet officials about one third were known or suspected Soviet intelligence officers. In one visit to a Boeing laboratory a delegate applied adhesive to his shoes to obtain metal samples. The size of the Soviet onslaught was so large that entire fields of US research and development became replicated in the Soviet Union. The East German spy master, Markus Wolf, recalled the East German computer company, Robotron, was, thanks to Soviet espionage, an unofficial subsidiary of IBM.

Soviet S&T espionage was often facilitated by sloppy security at US defense contractors. An employee of TRW Corporations in Redondo Beach, CA, which manufactured a US spy satellite, recalled that workers “regularly partied and boozed it up during working hours with the ‘black vault’ housing the Rhyolite [spy] satellite project”. Bacardi rum, he claimed, was kept behind the cipher machines and a cipher-destruction device was used as a blender to mix banana daiquiris and Mai-Tais.

Soviet espionage was so far reaching that, ironically, by the end of the Cold War both sides of the conflict, NATO and the Soviet Union, were dependent on US S&T.

Business Risk

Fast forward to the present day – a time when the world is vastly more complicated than the last century’s Cold War. Western countries did not need the Soviet economy. By contrast, China is intertwined with the world economy.

Beijing is seeking to portray Washington’s new approach to economic, defense, and foreign policies as undermining the post war rules based international order that it created. Meanwhile the PRC, which, has previously railed against the international order (though in reality it vastly benefited from that order), is holding itself out to be the stable player on the world stage. The PRC’s attitude is that it will play by the rules when it suits it but is happy to break them whenever it decides to do so.

Recent diplomatic outreach to Beijing by Western leaders include agreements framed as pragmatic economic wins. If middle powers, like Britain, for example, pursue a strengthening of commercial relations with China, they will need a strategy to mitigate risk like Nixon and Kissinger developed. Britain does not have a strategy for doing so. Even China hawks, like former US ambassador in Beijing, Nicholas Burns, have stated that for economic growth the US will need to continue to trade with China, but will need to carve out elements of national security and critical infrastructure. The latter is a principle stretching back to the UK government War Book.

There is no reason why the PRC would not seek to exploit a détente with the west as the Soviets did before. The Chinese state and its intelligence services have never encountered a western business whose intellectual property they did not want. The Chinese Communist Party (CCP) uses a constellation of front companies to do business with the outside world. Often such companies will enter into business ventures to obtain intellectual property from their western counterparts, but then pull the plug, bankrupting their western counter parties. To add insult to injury, Chinese firms will often sell the product they have stolen back to western markets.

The name of the game for western businesses must therefore be risk mitigation regarding China. In the last century, governments held the monopoly on the know-how and intelligence critical to the technologies that shaped our world – nuclear weapons. It took state resources to detect technology transfer. Soviet S&T espionage was only discovered when French intelligence recruited an agent in KGB Line-X in the 1980s. The same is not true today. Private sector companies today hold the keys to innovations that will shape our lives this century – microchips, A.I., quantum and bioengineering. It is therefore private sector companies that are best placed to mitigate risk of stolen intellectual property. And unlike in the past, this can be done by using A.I. driven publicly available data.

Have a perspective to share based on your experience in the national security field? Send it to Editor@thecipherbrief.com for publication consideration.

Read more expert-driven national security insights, perspective and analysis in The Cipher Brief

I just want to read the news

Get the news and only the news!

Guardian

Back to top

Kaupunki

Back to top

Yle

Back to top

CNN

Back to top

Hesari

Back to top

Al Jazeera

Back to top

New York Times

Back to top

Reuters

Back to top

NPR

Back to top

The Cipher Brief

Back to top