Organizers of a new initiative say governments, AI firms, security experts and the global epistemic community concerned with human survival must jointly shape technology before today’s choices become tomorrow’s constraints.
July/August 2026
By Douglas B. Shaw, Stephen Herzog, and William C. Potter
“All models are wrong, but some are useful,” observed statistician George E.P. Box.1 Indeed, today’s large language models and other artificial intelligence systems are far from perfect. Yet, as components of larger social and technical systems, these tools may be reaching a threshold at which their design could shape the future of world order as much as the choices of their users. Frontier AI laboratories have consequently become powerful actors, capable of altering global security faster than states and international organizations can design regulations.2 Decisions about whether and how to release advanced models can directly affect risks to humanity posed by weapons of mass destruction (WMD).
In response, the James Martin Center for Nonproliferation Studies launched the “Asilomar process for AI and global security” April 8-9 with a conference titled “Silicon, Swords, and Ploughshares: The Perils and Promise of AI in the Nuclear and Biological Domains.”3 The meeting in Pacific Grove, California, brought together more than 100 experts from universities, think tanks, governments, U.S. national laboratories, and the AI industry. The objective was to begin imagining a new chapter of anticipatory arms control necessitated by the collision of powerful AI tools with catastrophic dangers such as nuclear war, WMD proliferation, and pandemics.
The conference addressed a gaping disconnect between two ways of thinking about AI risks. One, often called AI “doomerism,” views AI as a possible source of human extinction or loss of control over our collective future. The other, a narrower misuse paradigm, asks whether low-capability actors could use AI to build a nuclear weapon, precipitate a meltdown of a nuclear power plant, or engineer a pathogen capable of causing a pandemic. But much practical work lies between these camps. Governments, AI firms, security experts, and the global epistemic community concerned with human survival must jointly shape technology before today’s choices become tomorrow’s constraints.
Why AI Needs Anticipatory Arms Control
Anticipatory arms control seeks restraint before a technology arrives in forms that are already militarized or difficult to restrict. AI fits poorly with the legacy of treaty-based constraints on capabilities that states can define, limit, and monitor. Its most important effects may appear through systems it changes rather than objects that can be counted or banned. In the nuclear and biological domains, AI could reshape proliferation pathways and elements of strategic stability.
This WMD-relevant challenge also arrives as arms control institutions are perhaps least able to absorb a fast-moving technological shock. The New Strategic Arms Reduction Treaty expired February 5 without a replacement. The 2026 nuclear Nonproliferation Treaty (NPT) Review Conference failed to achieve a consensus final document, and negotiators removed all language pertaining to AI from the draft text.4 Meanwhile, the Biological Weapons Convention (BWC) still lacks an effective verification mechanism.5
Changes in military innovation compound these difficulties.6 AI’s most consequential capabilities are emerging from frontier laboratories and startups, often before regulators even understand the problems and opportunities that they pose. The technology rewards speed, offering market incentives to labs and strategic advantages to states.7 That uncertainty and private sector acceleration make the current limitations of AI models potentially misleading.8 Complacency here would be ill advised, given the technology’s rapid trajectory.
Jane Vaynman and Tristan Volpe’s work on “dual-use deception” helps explain why formal AI arms control may be contentious.9 Restraint becomes harder when military and civilian applications are difficult to distinguish and a technology is deeply integrated into military systems and the civilian economy. AI has this character in spades. Threats posed by AI depend far less on the model itself than on its user, what information it can access, and what decisions it is allowed to shape.
The Asilomar process aims to encourage anticipatory arms control before the machinery of governance has caught up. It asks whether laboratories building frontier systems can sit with governments and weapons experts to shape choices that may become facts on the ground. Once model capabilities are released, adopted by nuclear planners, or woven into biological infrastructure, the space for restraint narrows considerably.
Principles for a Missing Forum
The process deliberately invokes a tradition of scientific self-governance. The 1975 Asilomar Conference on Recombinant DNA in Pacific Grove, California, addressed an emerging biological technology whose risks were still being defined.10 Later gatherings at the same site helped make “Asilomar” synonymous with scientists confronting the social consequences of their work.
The April 2026 meeting tested whether the right communities could reason together before a crisis. The process begins with the premise that AI developers, governments, and security experts each see only part of the picture. There is a missing standing forum where those partial perspectives can be bridged early enough to improve AI governance.
After further deliberation, the conference secretariat released seven principles for governing AI applications in nuclear and biological security, available in full elsewhere:11
Importantly, the principles cover both model-level choices inside AI laboratories and the responsibilities of states, international organizations, and the arms control community. They are meant to give these actors a common starting point to focus on restraint before emergent capabilities harden into crises.
Lessons for Nuclear and Biological Security
The gathering clarified that most AI-related nuclear and biological risks may lie between the poles of extinction speculation and the “bomb in a cave.” The latter phrase captures the fear that a terrorist group or technically weak state could use AI to leap over material and organizational barriers to building a weapon. That concern is legitimate, especially in synthetic biology, but it should not define the field of vision.
Material constraints to building and maintaining nuclear weapons are considerable: fissile materials, precision manufacturing, delivery systems, testing, and command and control. Recent research suggests, however, that AI could eventually help technically advanced would-be proliferators navigate bottlenecks to weaponization.12 Likewise, purpose-built AI might assist nuclear-armed states in designing more sophisticated warheads. AI integration into command-and-control or decision-support systems may also shape how states in deterrence relationships interpret warnings or assess adversarial behavior.13 The late Daniel Ellsberg described multiple large nuclear arsenals ready for prompt use as a single, globe-spanning “doomsday machine.”14 It might be triggered by any of several governments, but none can prevent this independently. AI could deepen that danger if it makes leaders believe they have found a fleeting advantage or a usable window for escalation. AI also might enable nonstate actors to identify vulnerabilities in civilian nuclear facilities, which could be exploited to unleash nuclear violence.15
Biological risks, made vivid by COVID-19, raise a different problem. Advanced AI systems may make biological design feel less like specialized knowledge and more like an assisted workflow, especially when paired with cloud laboratories that let remote users run experiments.16 No one should pretend that engineering a pandemic would be easy. Yet, those responsible for monitoring and verifying such events may increasingly not know where to look until risky work is already well underway.17
AI also could strengthen tools for evaluating compliance with agreements and norms without replacing human judgment.18 Machine learning applied to satellite imagery may help analysts notice changes at sensitive facilities.19 Data fusion across platforms could assist interpretation of noisy indicators, from procurement patterns and orders for biological materials or services to remote laboratory activity. Digital twins of sensitive facilities could help inspectors from the International Atomic Energy Agency and the Organisation for the Prohibition of Chemical Weapons prepare better questions before they arrive at sites.
Although not every catastrophic scenario involving advanced technology belongs in the same category, the lesson for nuclear and biological security is that AI may profoundly alter both landscapes long before the usual guardians of restraint know how the world has changed. The Asilomar process endeavors to look beyond weak actors seeking shortcuts and ask how powerful tools may shape decisions about WMD before new patterns become normalized.
Where Diplomacy Has Not Reached
The process is meant to strengthen, not sidestep, existing forums tasked with ensuring nuclear and biological restraint. The NPT and BWC remain indispensable, but neither is well-positioned to absorb the AI question quickly. Consensus diplomacy can discourage bold and timely action regarding emerging technologies, while the BWC’s absence of verification will become far more consequential as AI touches the practical work of synthetic biology. Concerningly, AI could potentially enable deadly biological capabilities to emerge before anyone can confidently say that a weapons program exists.
AI diplomacy has fortunately moved faster in some spheres than arms control diplomacy, although at a higher level of generality. The AI safety summits at Bletchley Park in 2023, Seoul in 2024, and Paris in 2025 drew international attention to frontier AI risk.20 The “responsible AI in the military domain” process—with summits in The Hague in 2023, Seoul in 2024, and A Coruña in 2026—has also done important work.21 These gatherings have a considerable international reach, but their breadth is also limiting. Nuclear and biological arms control can disappear inside larger conversations about AI safety, innovation, military autonomy, and responsible uses.
The Asilomar process is intended to fill that vital space. It focuses on catastrophic nuclear and biological weapons dangers, can move faster than treaty review processes, and aims to bring sometimes disparate expert communities into conversation.
AI’s ‘Pugwash’ Moment?
The April 2026 meeting was only the beginning. The process is an ongoing effort to support new standards of anticipatory arms control among frontier AI laboratories, governments, and security experts. But the next phase must expand the circle beyond participants already convinced that AI will shape global security. Many firms, investors, and research communities driving humanity’s AI future do not yet see nuclear and biological restraint as their responsibility. Many also harbor heroic expectations about the capacity of governments to regulate exquisite tools still being imagined inside the private sector. The time to develop shared AI standards and thresholds in the nuclear and biological domains has already arrived, while oversight is dramatically lagging.
To their credit, some forward-leaning frontier labs have begun hiring staff to think seriously about security consequences of their most powerful models, but the pool of qualified talent is shallow and there is great competition for those individuals. Some firms also have experimented with more restrictive model release practices.22 As a result, several leading AI companies already have endorsed the need for an ongoing Asilomar process, as have multiple states. They share a recognition that the primacy of state governance on issues involving dual-use disruptive technologies has diminished and is unlikely to change trajectory. As such, early engagement between industry and government offers the best prospects for ensuring the development of technologically sound rules and practices that serve both commercial and national security interests.
Even with deeper industry engagement, the participant base must include intergenerational voices from around the world, a point underscored by India’s hosting of the 2026 AI Impact Summit in New Delhi.23 A process shaped predominantly by powerful Western states and their best-capitalized companies would lack legitimacy and likely reproduce inequalities that have long undercut nuclear and biological governance. Those who bear the consequences of new rules should have a role in shaping them.
A second Asilomar conference is planned for April 2027—by the James Martin Center in partnership with the UN Institute for Disarmament Research and the Berkeley Risk and Security Lab—to deepen and widen this discussion. The process takes inspiration from the Pugwash Conferences on Science and World Affairs, founded in 1957 and later awarded the Nobel Peace Prize for helping scientists pull the world back from the nuclear brink.24 There is reason for cautious optimism that this process can awaken similar awareness and collaboration among those who shape transformational AI technologies.
ENDNOTES
1. George E.P. Box, “Robustness in the Strategy of Scientific Model Building,” in Robustness in Statistics: Proceedings of a Workshop, eds. Robert L. Launer and Graham N. Wilkinson (New York: Academic Press, 1979), p. 202.
2. On this “pacing problem,” see Gary E. Marchant, Braden R. Allenby, and Joseph R. Herkert, eds., The Growing Gap Between Emerging Technologies and Legal-Ethical Oversight: The Pacing Problem (Dordrecht, Netherlands: Springer, 2011); Justin Key Canfil, “Convergent Flexibility: How International Law Keeps Pace with Technological Change,” International Organization, Vol. 80, No. 2 (Spring 2026): pp. 237-280.
3. James Martin Center for Nonproliferation Studies, “CNS Organizes Iconic Asilomar Conference on AI and WMD,” April 14, 2026.
4. See “Implementation of the Treaty on the Non-Proliferation of Nuclear Weapons for its Twelfth Review Cycle,” 2026 Review Conference of the Parties to the Treaty on the Non-Proliferation of Nuclear Weapons, NPT/CONF.2026/CRP.4, May 21, 2026.
5. Matthew P. Shearer, Christina Potter, Rachel A. Vahey, Nancy D. Connell, and Gigi Kwik Gronvall, “BWC assurance: increasing certainty in BWC compliance,” The Nonproliferation Review, Vol. 29, Nos. 1-3 (February-June 2022): pp. 47-75.
6. Stephen Herzog and Dominika Kunertova, “NATO and Emerging Technologies—The Alliance’s Shifting Approach to Military Innovation,” Naval War College Review, Vol. 77, No. 2 (Spring 2024): pp. 47-69.
7. Michael C. Horowitz, “Artificial Intelligence, International Competition, and the Balance of Power,” Texas National Security Review, Vol. 1, No. 3 (May 2018): pp. 36-57.
8. David M. Allison, “Integrating Artificial Intelligence with Nuclear Weapons: A New Paradigm of Irreversible Transformation,” The Nonproliferation Review, Vol. 33, Nos. 1-3 (2026): forthcoming.
9. Jane Vaynman and Tristan A. Volpe, “Dual Use Deception: How Technology Shapes Cooperation in International Relations,” International Organization, Vol. 77, No. 3 (Summer 2023): pp. 599-632.
10. Paul Berg, David Baltimore, Sydney Brenner, Richard O. Roblin III, and Maxine F. Singer, “Summary Statement of the Asilomar Conference on Recombinant DNA Molecules,” Proceedings of the National Academy of Sciences, Vol. 72, No. 6 (June 1975): pp. 1,981-1,984.
11. Stephen Herzog, Allison Berke, Yanliang Pan, William C. Potter, and Douglas B. Shaw, “AI is changing biological and nuclear risks; governance must change accordingly,” Bulletin of the Atomic Scientists, June 18, 2026.
12. David M. Allison and Stephen Herzog, “Artificial Intelligence and Nuclear Weapons Proliferation: The Technological Arms Race for (In)visibility,” Risk Analysis, Vol. 45, No. 11 (November 2025): pp. 3,839-3,859.
13. Herbert S. Lin, “Artificial Intelligence and Nuclear Weapons: A Commonsense Approach to Understanding Costs and Benefits,” Texas National Security Review, Vol. 8, No. 3 (Summer 2025): pp. 98-109; Douglas B. Shaw, Isabelle Williams, Patricia Jaworek, Kevin Park, and Pravin Rajan, “Nuclear Security Implications of AI and Emerging Technologies: A FutureSafe Analysis of Risks and Opportunities,” Nuclear Threat Initiative, November 13, 2025; and Joshua A. Schwartz and Michael C. Horowitz, “Delegating Destruction: Coercive Threats and Automated Nuclear Systems,” International Organization, Vol. 80, No. 1 (Winter 2026): pp. 179-204.
14. Daniel Ellsberg, The Doomsday Machine: Confessions of a Nuclear War Planner (New York: Bloomsbury, 2017).
15. Sarah Case Lackner and Zaheed Kara, “Artificial Intelligence and Nuclear Security Governance: Addressing the Risks of Frontier AI,” Vienna Center for Disarmament and Non-Proliferation, December 2025; Muhammed Ali Alkış, Zoha Naser, and Sarah Tzinieris, “A Fifth Face? The Evolving Threat of Nuclear Terrorism in the Age of Artificial Intelligence,” The Nonproliferation Review, Vol. 33, Nos. 1-3 (2026): forthcoming.
16. Jeffrey Lee, Bria Persaud, Barbara Del Castello, Allison Berke, and Gustavs Zilgalvis, “Documenting Cloud Labs and Examining How Remotely Operated Automated Laboratories Could Enable Bad Actors,” RAND Corporation, PE-A3851-1, April 2025.
17. Doni Bloomfield et al., “AI and biosecurity: The need for governance,” Science, Vol. 385, No. 6,711 (August 22, 2024): pp. 831-833.
18. Jane Vaynman, “Better Monitoring and Better Spying: The Implications of Emerging Technology for Arms Control,” Texas National Security Review, Vol. 4, No. 4 (Fall 2021): pp. 33-56; Francisco Parada, Nathan Martindale, Alisa Reasor, Scott Stewart, and Lindsey Ukishima, “AI for Nuclear Safeguards Verification,” Oak Ridge National Laboratory, ORNL/LTR-2024/3674, October 2024; OPCW temporary working group on AI, “Artificial Intelligence: Report of the Scientific Advisory Board’s Temporary Working Group,” Organisation for the Prohibition of Chemical Weapons, SAB/REP/1/26, March 2026.
19. Yanliang Pan, “Next-Generation OSINT: Machine Learning for Tracking Nuclear and Missile Proliferation,” The Nonproliferation Review, Vol. 33, Nos. 1-3 (2026): forthcoming.
20. Tony Oweke, “The World Is Trying to Govern AI. The UN Wants In.” Council on Foreign Relations, May 29, 2026.
21. Giacomo Persi Paoli and Yasmin Afina, “AI in the Military Domain: A briefing note for States,” United Nations Institute for Disarmament Research, March 2025; Jules Palayer and Laura Bruun, “Artificial intelligence and international peace and security,” in SIPRI Yearbook 2025: Armaments, Disarmament and International Security (Oxford University Press, 2025), pp. 329-346.
22. See, for example, Anthropic Frontier Red Team, “Assessing Claude Mythos Preview’s cybersecurity capabilities,” Anthropic, April 7, 2026.
23. Indian Ministry of Electronics and Information Technology, “India AI Impact Summit 2026,” February 16-20, 2026.
24. John P. Holdren, “Arms Limitation and Peace Building in the Post-Cold-War World,” Nobel lecture on behalf of the Pugwash Conferences on Science and World Affairs, Oslo, December 10, 1995.
Douglas B. Shaw is a nonresident scholar at the James Martin Center for Nonproliferation Studies at the Middlebury Institute of International Studies at Monterey. Stephen Herzog is the James Martin Center’s professor of the practice and an associate of the Project on Managing the Atom at Harvard’s Belfer Center for Science and International Affairs. William C. Potter is the director of the James Martin Center and its Sam Nunn and Richard Lugar professor of nonproliferation studies.
India Commissions Third Ballistic Missile Submarine
July/August 2026
India commissioned its third nuclear-powered ballistic missile submarine (SSBN), the Arihat-class INS Aridhaman.
Indian media reported that sources within India’s defense and security establishment confirmed the April 3 commissioning, although the government has not officially announced it. Defense Minister Rajnath Singh did, however, post in Hindi on social media April 3, that “It’s not words but power, ‘Aridhaman’!”
Possessing three operational SSBNs is a crucial threshold for India because “To ensure round-the-clock operational deployment in the ocean, at least three submarines are required—one on patrol and the other two in maintenance or transit,” Dinakar Peri, a fellow at the Carnegie Endowment for International Peace, wrote April 24 in an analysis.
The INS Aridhaman also has several technological advances that enhance India’s retaliatory abilities as it seeks to defend against China and Pakistan. These include the ability to carry the K-4 submarine-launched ballistic missile, with a 3,500-kilometer range, once that missile completes its testing phase. The ship has eight vertical launch tubes, twice as many as in the first two submarines in its class.
India has already launched the next Arihat-class submarine, which experts expect to be commissioned next year.
According to the Stockholm International Peace Research Institute’s 2026 Yearbook, India may have begun deploying a small number of warheads on one of its SSBNs for peacetime deterrence patrols during 2025. The authors assign “considerable uncertainty” to their assessment that 12 warheads are deployed in this mode. Any coupling of missiles with their launchers during peacetime would be a departure from previous Indian policy.
—MIA CLARKE
July/August 2026 Digital Magazine
Iran and the United States announced a memorandum of understanding to end their war and begin new talks on a nuclear agreement, but negotiators face significant obstacles to a final deal.
July/August 2026
By Kelsey Davenport
Iran and the United States announced a memorandum of understanding to end their war and begin new talks on a nuclear agreement, but negotiators face significant obstacles to a final deal, including a resurgence in fighting.
Pakistani Prime Minister Shehbaz Sharif announced June 14 on the social media platform X that the United States and Iran reached a peace agreement and that, “Both sides have declared the immediate and permanent termination of military operations on all fronts, including in Lebanon.”
The ceasefire broke down two weeks later when the United States blamed Iran for launching drone attacks on two vessels in the Strait of Hormuz and retaliated. Despite the strikes, the U.S. and Iranian negotiating teams conducted a second round of talks on implementing the memorandum in Doha June 30-July 1.
The United States and Israel initiated the 2026 war when they illegally attacked Iran Feb. 28, prompting retaliatory strikes by Iran that included targets in Gulf states. Several days later, Israel began attacking targets in Lebanon that it claims are tied to the Iranian-supported Hezbollah militia.
Pakistan helped broker an initial ceasefire between the United States and Iran April 8 and mediated talks on the memorandum.
Following Sharif’s announcement, U.S. President Donald Trump confirmed the deal in a June 14 post on Truth Social saying “Congratulations to all!” He announced that the Strait of Hormuz, a critical waterway for shipping oil and other commodities that Iran closed by targeting ships, was now open as a result of the memorandum.
Iranian Supreme Leader Ayatollah Mojtaba Khamenei said in a June 18 message shared with Iranian media that he “issued [his] permission” for Iran to sign the memorandum despite having a “different view” of what should be included. He warned that “in-person negotiations that will take place in the future will not mean acceptance of the enemy's position.”
Trump, in Versailles, and Iranian President Masoud Pezeshkian, in Tehran, signed the memorandum June 17. The next day, Pezeshkian called the memorandum an “historic document.” He wrote on X that “peace will be achieved in the shadow of mutual respect.”
The United States released the text of the 14-paragraph memorandum June 17. It said that the United States, Iran, and their allies “declare the immediate and permanent termination of military operations on all fronts, including in Lebanon, and undertake from now on not to initiate any war or military operation against each other, and to refrain from the threat or use of force against each other.”
Specifically on the nuclear program, the memorandum reaffirmed Iran’s commitment to never “procure or develop nuclear weapons.” It also noted that the United States and Iran “agreed to resolve the disposition of stockpiled, enriched material pursuant … with the minimum methodology to be down-blending on site” under International Atomic Energy Agency (IAEA) supervision. The issue of uranium enrichment and other nuclear issues would be negotiated as part of a final deal, according to the memorandum. In return, the United States committed to “terminate all types of sanctions against Iran,” including at the United Nations. It specified that U.S. primary and secondary sanctions on Iran would be included in the sanctions lifting.
The memorandum committed the United States and Iran to negotiate a final nuclear deal “in maximum 60 days extendable with mutual consent.”
The agreement declared the immediate reopening of the Strait of Hormuz and committed Iran and Oman to “define the future administration and maritime services” through that waterway, in discussion with other regional states. Before the war, the Strait of Hormuz was an open waterway, but Iran now seeks to profit from it.
The memorandum includes a commitment to establish a fund to rebuild Iran and requires that the United States issue waivers for Iranian oil sales. The U.S. Treasury Department quickly issued a license for Iran to sell oil and other petrochemical products through Aug. 21.
Although Vice President JD Vance, who was deeply involved in the memorandum negotiations, repeatedly stated that Iran does not receive any unilateral benefits, the upfront sanctions relief and vague nuclear commitments spurred bipartisan criticism of the memorandum as a win for Iran.
Reaching a final agreement in 60 days will be challenging, as negotiators appear far apart on key issues and will likely face opposition.
Although the memorandum references Iran, the United States, and “their allies,” Israeli Prime Minister Benjamin Netanyahu said his country is not bound by the interim deal. He said June 14 that Israeli troops “will remain in the security zone in southern Lebanon for as long as necessary” to protect against Hezbollah attacks. Netanyahu also continues to reiterate that Iran must give up its entire nuclear program as part of any agreement.
Trump wrote June 18 on Truth Social that he expects a “complete Ceasefire on all fronts” and that all states in the Middle East should “maintain their commitment to allowing our negotiations to beautifully unfold.” He has also threatened to resume attacks on Iran if a deal is not reached, in an apparent violation of the ceasefire.
Despite Trump’s threat, U.S. and Iranian negotiating teams met in Lucerne, Switzerland, June 21-22 for the first round of talks after reaching the memorandum.
Pakistan and Qatar mediated the talks and afterwards said that “encouraging progress has been made, including the creation of a mechanism for further technical talks.” A high-level committee will provide political oversight of the process and there will be working groups on nuclear issues, sanctions, monitoring, and dispute resolution, according to a June 22 statement.
After the talks, Iranian Foreign Minister Abbas Araghchi said on X that they “delivered major progress to end Lebanon War.” He described deconfliction in Lebanon as the first “real test” of the process.
The two sides offered different accounts of what transpired on the nuclear issue. In a press conference following the meetings, Vance said that “we made a lot of good progress on other nuclear talks.”
Iranian Foreign Ministry spokesman Esmail Baghaei said that the United States presented its “positions regarding the nuclear issue,” but there was “no discussion of details.” He suggested that nuclear issues also were not a focus of the second round of talks in Doha.
Trump suggested to The New York Times in a June 14 interview that the United States would accept a 15-year suspension on Iran’s uranium enrichment, moving away from his previous demand for a permanent enrichment ban. Iran is open to a suspension but seeks a shorter duration. Although the memorandum suggests that the United States is now open to diluting the highly enriched uranium in Iran, it still appears that Trump wants the blended-down material shipped elsewhere, whereas Iran wants to keep the enriched uranium in the country.
Before the Lucerne talks, Vance suggested that the two sides would discuss a resumption of IAEA inspections. Iran is legally required to implement a safeguards agreement with the IAEA, but it suspended cooperation with the agency and blocked its access to bombed sites after the 12-day U.S.-Israeli war in June 2025. (See ACT, July/August 2025.)
After the meetings ended, Vance said the United States, the IAEA, and Iran will discuss inspections “this week.”
Iranian Deputy Foreign Minister Kazem Gharibabadi suggested the resumption of inspections is not imminent. He said in a June 24 post on X that Iran held “no meeting with [IAEA Director-General Rafael Mariano] Grossi took place in Switzerland, despite his request,” and that there is “no plan for access to the attacked facilities and nuclear materials.” Inspections will be “resolved with the framework of the final agreement,” he said.
Grossi told reporters at a June 24 press conference in Japan that “inspections will indeed take place.” The IAEA and Iran will be “working on the modalities—dates, procedures, places—very soon,” he said.
The new rules would impose a regulatory framework for autonomous weapons and military artificial intelligence that emphasizes human judgment and ultimate human responsibility, while also endorsing the Pentagon’s embrace of the technology.
July/August 2026
By Xiaodon Liang
A key U.S. Senate committee recommended imposing on the military a regulatory framework for autonomous weapons and military artificial intelligence that emphasizes human judgment and ultimate human responsibility, while largely endorsing the Pentagon’s embrace of lethal autonomous weapons and urging that the department “maximize uses” of these emergent technologies.
The Senate Armed Services Committee endorsed the new framework in its draft of the fiscal year 2027 National Defense Authorization Act, which the panel finished marking up June 10.
The sections pertaining to autonomous weapons and military AI would require the department to “ensure personnel exercise appropriate levels of human judgment” and that relevant systems are “designed and employed in a manner that enables commanders and operators to exercise ultimate human responsibility over the use of force.”
The requirement of “appropriate levels of human judgment” is already established in Department of Defense Directive 3000.09, on autonomy in weapons systems, but has lacked a clear definition. (See ACT, March 2023.)
The new framework seeks to elaborate on the principle by specifying that systems must allow for supervision by human operators, include methods for intervention or termination, include fail-safe mechanisms to enable manual control, provide adequate monitoring data to controllers, maintain records of target selection data and logic, and operate in compliance with “international law, rules of engagement, the law of armed conflict, applicable treaties, and Department of Defense policy.”
The proposed legislation would codify the existing review process mandated by the 2023 directive, while also specifying verification, validation, testing, and evaluation criteria. The department’s office of the Director, Operational Test and Evaluation, which has had its proposed budget slashed by two-thirds in this year’s Pentagon budget request, would be charged with ensuring that covered systems function “as anticipated in realistic operational environments against adaptive adversaries and are sufficiently robust to minimize failures.”
The committee text also would establish an incident repository for tracking data on system failures, unintended behavior, and near-miss events.
The language on autonomy and AI would categorically prohibit the use of these technologies for any “decision to initiate the launch or detonation of a nuclear weapon,” a move sought on a bipartisan basis in 2023 by Sen. Ed Markey (D-Mass.), Rep. Ted Lieu (D-Calif.), Rep. Don Beyer (D-Va.) and Rep. Ken Buck (R-Colo.).
The new Senate Armed Services Committee proposal does not, however, include a ban on the use of AI for the targeting of nuclear weapons, as proposed by Sen. Kristen Gillibrand (D-N.Y.) in a stand-alone bill introduced June 2.
Gillibrand’s bill was one of several introduced in the last few weeks seeking to address autonomous weapons and military use of AI by primarily Democratic members of Congress. The others included separate proposals by Sen. Adam Schiff (D-Calif.), Sen. Elissa Slotkin (D-Mich.), and Sen. Mark Kelly (D-Ariz.) in collaboration with Rep. Suhas Subramanyam (D-Va.).
Ultimately, the Senate committee language largely tracks a bill introduced June 8 by Democratic leadership, represented by ranking member Jack Reed (D-R.I.) and his counterpart on the Senate’s defense appropriations subcommittee, Chris Coons (D-Del.). The text, which was accepted by the Republican majority into the base text of the draft of the key policy bill, also incorporates substantial portions of the Kelly-Subramanyam bill.
The Senate committee text, as based on the Coons-Reed bill, is less specific in its protections against domestic surveillance as compared to the other proposals put forward by Democratic senators.
Elsewhere in the draft bill, the committee recommends that the Senate mandate creation of a robotics and autonomous systems command that would be responsible for “force generation, joint training, interoperability, doctrine development, and operational employment through other combatant commands.”
Reflecting bipartisan concerns about the effects of AI adoption, a separate section in the Senate’s draft act incorporates a bill proposed by Kelly and Sen. Tom Cotton (R-Ark.) that would instruct the Pentagon to assess the effect of AI use “on the maintenance and retention of essential warfighter skills.”
Otherwise, the text is largely supportive of Pentagon efforts to expand usage of generative AI and so-called agents, or AI-powered autonomous software.
The new draft legislation for regulation of autonomous weapons systems and military AI faces an uncertain path toward gaining final passage by the full Congress.
The sparer House variant of the defense policy bill, which was approved by the House Armed Services Committee June 5, instructs the Pentagon to update its directive on autonomous systems and military AI while taking into account many of the issues covered in the Senate bill, but provides less specificity and more deference to the Pentagon.
The recent wave of legislative interest in the Senate in autonomous weapons and military AI reflects growing unease over the Trump administration’s revisions to Pentagon policies on the use of AI.
Secretary of Defense Pete Hegseth instructed the department Jan. 9 to adopt a new policy that vendors be required to accept that their AI models may be deployed for “any lawful use.” The order implied that previous departmental restraints would be reassessed.
In February and March, the Pentagon took unprecedented coercive actions against a prominent U.S. technology firm, Anthropic, when the company rejected contract modifications that removed safeguards against use of its AI models in fully autonomous weapons systems and for domestic mass surveillance. (See ACT, April 2026.)
President Donald Trump signed a national security presidential memorandum on AI in the national security enterprise June 5 that, among other measures, tasks the Pentagon to rewrite Directive 3000.09, which was last updated in January 2023.
U.S. Defense Secretary Pete Hegseth announced that the six-month review of U.S. basing and force posture in Europe was aimed at preparing for “NATO 3.0.”
July/August 2026
By Libby Flatoff
U.S. Defense Secretary Pete Hegseth announced that the United States will conduct a six-month review of U.S. basing and force posture in Europe for what he called “NATO 3.0.”
“[W]e’re doubling down on our effort to make NATO what it always was supposed to be, a balanced alliance with Europe in the lead for its own defense,” Hegseth told NATO defense ministers in Brussels June 18.
The announcement came after The Financial Times reported June 1 that the United States was in internal discussions with the alliance about deploying nuclear dual-capable aircraft to additional countries in Europe.
Belgium, Germany, Italy, the Netherlands, and Turkey currently operate their own dual-capable aircraft, purchased from the United States and designed and certified to deliver U.S. nuclear weapons. Turkey remains in a reserve and contingency role.
The newspaper said that two sources briefed on the discussions believe that this signaling was “intended to show the U.S. commitment to providing a nuclear umbrella even as NATO allies were pushed to shoulder more of the conventional defence burden.”
“Europe can and must take primary responsibility for its conventional defense as it pledged at the Hague Summit,” Hegseth told the defense ministers. “Going forward, [U.S.] annual NATO dues will be contingent on other countries meeting their defense spending targets. Where other allies do not spend with urgency, our dues contributions will go down. NATO will be a two-way street.”
On the same day, the Nuclear Planning Group, NATO’s senior body for nuclear deterrence, met and released a statement on nuclear deterrence matters for the first time since 2007. The group, which includes all NATO members except France, agreed to “continue enhancing NATO’s nuclear deterrence mission by modernising NATO’s nuclear capabilities, strengthening its nuclear planning capacity, and adapting to achieve its security interests.”
PBS News reported June 3 that Poland and Lithuania confirmed that they began discussions with the United States on their role within NATO nuclear deterrent efforts. Polish Deputy Defense Minister Pawel Zalewski told Polish Radio, “We are talking, in order to create better conditions for nuclear deterrence and for Poland to play an important role in that.”
Poland has also expressed interest in France’s new “forward deterrence” plans, which were announced by President Emmanuel Macron in March and could allow for the temporary deployment of nuclear strike aircraft outside of France. (See ACT, March 2026.) Belgium, Denmark, Germany, Greece, the Netherlands, Sweden, and the United Kingdom have expressed their intention to discuss nuclear cooperation with France.
Most recently, Norway announced in a May 27 press release that it signed a comprehensive defense agreement with France “to commit to come to each other’s aid should the need arise.” The press release added that, “Together with nine other European countries, Norway will also participate in a process to determine how France’s nuclear weapons can contribute to security and deterrence in Europe.”
Finland, which joined the alliance in 2023, lifted its national ban on nuclear weapons June 17, 2026, after a vote by parliament. Heikki Autto, chair of the parliamentary defense committee, stated in an interview that “nuclear deterrence is, in the end, the guarantor of peace in Europe.… It’s the ultimate deterrence against the Russian aggression.”
Russian spokesman Dmitry Peskov warned June 3 that, “By placing nuclear weapons on its territory, Finland will begin to pose a threat to us. And if Finland threatens us, we will take appropriate measures.”
In a March 5 press release, the Finnish Defense Ministry first announced its intention to amend the state’s Nuclear Energy Act to allow nuclear arms within its territory “to remove legal barriers to enable Finland’s homeland defence as part of the Alliance and the full utilisation of NATO’s deterrence and defence.” (See ACT, April 2026.)
The next NATO summit will be July 7-8 in Ankara.
State-run media said North Korean leader Kim Jong Un inspected the factory and reiterated the necessity of “implementing ambitious future plans designed to strengthen our state’s nuclear forces exponentially.”
July/August 2026
By Kelsey Davenport
North Korean leader Kim Jong Un visited a new uranium enrichment facility at an undisclosed location, which he described as necessary to bolster the country’s nuclear arsenal.
The state-run Rodong Sinmum published June 3 a report of Kim’s visit to a “newly commissioned nuclear material production plant” and showed a photo of him walking between rows of centrifuges, machines used to enrich uranium, at the site.
The report said that Kim inspected the factory and reiterated the necessity of “implementing ambitious future plans designed to strengthen our state’s nuclear forces exponentially.” Kim expressed satisfaction with the facility and said that North Korea’s “weapons-grade nuclear materials production capacity more than doubled during the past 5-year” period.
North Korea announced plans to continue expanding its nuclear arsenal at the Ninth Party Congress, which took place in February. (See ACT, March 2026.)
After the visit, Kim held a “consultative meeting on bolstering the nuclear forces,” according to the report. At the conclusion of the meeting, Kim announced that North Korea “updated the digits that are critical for our nuclear activities.”
The updated number “comprehensively considered the tactical and strategic requirements for building up our nuclear deterrent,” the report said, suggesting that Kim set a new target for North Korea’s stockpile of nuclear warheads. Fissile material production capacity must continue to grow to increase the country’s nuclear arsenal, Kim was quoted as saying.
North Korea does not disclose the size of its nuclear arsenal, but a March 2026 report from the Federation of American Scientists estimated that North Korea possesses about 60 nuclear warheads.
Rodong Sinmum and other state-run media did not disclose the location of the facility, but it is likely within the Yongbyon nuclear complex.
The International Atomic Energy Agency (IAEA) has been tracking the construction of a building at the site that is similar in size and design to two other known uranium enrichment facilities in North Korea. Satellite imagery indicates that North Korea began construction on the building in December 2024.
In a June 8 statement to the agency’s Board of Governors, IAEA Director-General Rafael Mariano Grossi referenced North Korea’s reports on the new enrichment facility and said that the “interior structure and layout of the facility is consistent with the Agency’s observation of the features of the new building at Yongbyon.”
He said that the “ongoing operation of enrichment facilities at Kangson and Yongbyon and the reported further expansion of [North Korea’s] production of weapons-grade nuclear material is a cause for serious concern.”
The June 3 report from Rodong Sinmum suggested that the centrifuges at the newly unveiled facility will enrich uranium more efficiently because the new site features “more sophisticated technologies.”
In addition to the expanding uranium enrichment capacity, Grossi noted that it appears that North Korea is continuing to operate the 5-megawatt reactor at Yongbyon. North Korea uses plutonium separated from the spent fuel produced by that reactor for nuclear weapons.
In a June 10 statement to the IAEA Board of Governors, Howard Solomon, the chargé d’affaires at the U.S. mission to international organizations in Vienna, said that the United States is “gravely concerned by [North Korea’s] ongoing nuclear activities” and described reports on the new enrichment facility as “deeply troubling.” The United States remains committed to the “complete denuclearization” of North Korea, he said.
North Korea continues to reject denuclearization and reaffirm its status as a nuclear-armed state. In a June 13 statement, the state-run Korean Central News Agency reported that the United States and South Korea are “accelerating arms buildup” in the region and reiterated that its nuclear arsenal is an “irreversibly finalized matter.”
The statement was issued after the United States and South Korea held the sixth meeting of the Nuclear Consultative Group, a body formed in 2023 to strengthen the alliance and provide Seoul with more input in Washington’s extended deterrence planning. (See ACT, May 2023.)
After the June 11 meeting, the United States in a joint statement “reaffirmed the commitment to provide extended deterrence to [South Korea], utilizing the full range of U.S. capabilities, including nuclear.”
A second statement published in KCNA June 13 criticized South Korean President Lee Jae Myung for condemning North Korea’s “illegal military cooperation” with Russia in a June 10 joint statement issued after the South Korea-European Union summit. The statement also said that North Korea “will never be accepted as a nuclear-weapon state.”
The North Korean Foreign Ministry said that Lee’s proposal for peaceful coexistence between the two Koreas is impossible. The statement said Lee’s push for peace was a “deceptive farce” and South Korea is an “enemy state obsessed with hostility and confrontation.”
During his trip to Europe, Lee met with President Donald Trump on the sidelines of the Group of 7 summit in France. Lee told reporters at a June 18 press briefing that he asked Trump to focus on North Korea after reaching a deal with Iran. Lee said he suggested a phased approach that leads to denuclearization.
The government said the nuclear-powered attack submarines will run on low-enriched fuel produced domestically with the first submarine launch expected in the mid-2030s.
July/August 2026
By Kelsey Davenport
South Korea’s nuclear-powered attack submarines will run on low-enriched fuel produced domestically, according to a government document outlining Seoul’s plans for the submarine program.
The Ministry of National Defense released the Basic Plan for the Development of the Republic of Korea Nuclear-Powered Submarine May 26. Defense Minister Ahn Gyu-back told press during the plan’s unveiling that South Korea aims to launch its first submarine in the mid-2030s.
The “nuclear-powered submarines can monitor and track North Korean submarine forces more covertly and swiftly than conventional diesel submarines,” Ahn said, thus significantly contributing to South Korea’s “underwater kill chain capability.” The underwater kill chain refers to South Korea’s plans to track and target North Korea’s nuclear-armed submarines before Pyongyang can attack.
The plan was published six months after President Donald Trump pledged U.S. support for South Korea’s decision to build nuclear-powered attack submarines and its ambition to enrich uranium and separate plutonium, capabilities Washington previously sought to prevent Seoul from acquiring. (See ACT, December 2025.)
The plan states that, “to secure autonomy and stability in force acquisition, sustainment, and maintenance, the nuclear-powered submarine will be developed and built within the Republic of Korea.”
The plan did not specify the uranium enrichment level for the planned submarine reactor fuel, but the document’s reference to “low-enriched” suggested that the fuel would be less than 20 percent enriched uranium-235. Uranium enriched to less than 20 percent poses less of a proliferation risk because it cannot immediately be used for nuclear weapons. But as a practical matter, if South Korea is producing the enriched uranium to fuel the reactors, it could use the same technology to produce weapons-grade uranium, which is enriched to above 90 percent U-235.
The submarine plan committed South Korea to “establish a safeguards system applicable to nuclear-powered submarines in cooperation” with the International Atomic Energy Agency (IAEA). Seoul also pledged to “faithfully fulfill its nuclear non-proliferation obligations throughout the entire process of securing and managing low-enriched uranium, the nuclear fuel required for the propulsion system of nuclear-powered submarines, in close communication with the United States.”
In a June 8 statement, IAEA Director-General Rafael Mariano Grossi said the agency noted South Korea’s commitment to “fulfilling its non-proliferation obligations” and welcomed its “intention to work with the Agency on safety, security and safeguards measures applicable to nuclear-powered submarines.”
After publication of the plan, South Korea and the United States met to discuss the nuclear cooperation provisions that Trump and Lee agreed to in November.
First Vice Foreign Minister Park Yoon-joo led the South Korean negotiating team. After the June 2-3 meeting with U.S. Undersecretary of State for Political Affairs Allison Hooker, Park said that a “separate agreement” with the United States on the reactor fuel will be necessary, because “nuclear-powered submarines are related to the military use of nuclear energy.” South Korea has a nuclear cooperation agreement with the United States, but it is restricted to civil nuclear cooperation.
Hooker described the discussions as a “success” in a June 3 post on the social media site X and said the talks “set the stage for continued technical conversations.”
Talks between Park and Hooker also covered enrichment and reprocessing under the existing 2015 civil nuclear cooperation agreement, which allows a high-level bilateral commission tasked with identifying “appropriate options” for South Korean domestic uranium enrichment up to 20 percent U-235. South Korea claims it needs enrichment and reprocessing to support its civil nuclear industry.
A top UN disarmament official told the UN Security Council that the cache included rockets of the same type used in the 2013 sarin gas attack in Ghouta that killed an estimated 1,400 people.
July/August 2026
By Libby Flatoff
Inspectors from the Organisation for the Prohibition on Chemical Weapons (OPCW) uncovered a significant cache of undeclared chemical weapons and documentation in Syria after the new government of Syrian President Ahmad al-Sharaa pledged to help the agency resolve outstanding issues relating to the arsenal amassed by deposed leader Bashar al-Assad.
In a briefing June 4, Undersecretary-General and High Representative for Disarmament Affairs Izumi Nakamitsu told the UN Security Council that the cache included rockets of the same type used in the 2013 sarin gas attack in Ghouta that killed an estimated 1,400 people. “These findings are a momentous discovery, not just for Syria, but for international security and the global disarmament regime,” she said.
According to a UN press release, Syria’s UN ambassador, Ibrahim Olabi, described the discoveries as a “decisive turning point” and a “major leap forward in delivering accountability.” He said that his government had facilitated 32 visits by OPCW inspectors and handed over more than 60,000 pages of documents.
The OPCW has been unable to confirm Syria’s chemical weapon declaration since it was submitted in 2014 by Assad’s government, which was ousted in 2024 after a civil war.
In 2017, Damascus used sarin to attack civilians in Khan Shaykhun. In April 2018, chlorine gas was used in Douma. The United States responded with airstrikes both times.
Mohamad Katoub, Syria’s ambassador to the OPCW, told Arms Control Today in May that it was important to confirm and take accountability for these stockpiles because, “Accountability is a measure of nonrecurrence.” (See ACT, May 2026.) “The Syrian people suffered from those chemical weapons, and part of justice is the destruction of the remnants of chemical weapons,” he said. “Without holding perpetrators accountable and bringing justice for the victims, the peace in Syria will keep being fragile.” However, Katoub acknowledged that he doesn’t “believe that we’ll be able to have a 100-percent accurate declaration.” He said that ferreting out remnants of the CW arsenal had been complicated partly by “the Israeli airstrikes [and] what’s happening now in the region.” He noted that the OPCW Technical Secretariat team had to be evacuated from Syria March 17 over security concerns after the U.S.-Israeli war against Iran began Feb. 28.
After the 2024 collapse of the Assad regime, Israel undertook several airstrikes against known Syrian chemical weapon stockpiles.
According to Israeli Foreign Minister Gideon Saar, “We attacked strategic weapons, the residual chemical weapons capabilities, [and] long-range missiles and rockets, so they won’t fall into the hands of radicals.”
In December 2024, the OPCW affirmed its “commitment to clarifying gaps, discrepancies, and inconsistencies in Syrian chemical weapons declaration amidst political transition.”
The UN Security Council press release said that Nakamitsu “stressed that the newly discovered weapons must now be formally declared and destroyed under OPCW verification, and that further inspections of additional sites are needed.”
During the Security Council briefing, France cautioned that this was “only just beginning to reveal the extent of the programme hidden from the international community.”
In the interview with ACT, Katoub said Syria is working to safeguard the new sites until the proper steps toward destruction can be done.
Under the unprecedented program, participating companies would receive weapons-grade plutonium recycled from U.S. nuclear weapons for use in nuclear power plants.
July/August 2026
By Mia Clarke
The Department of Energy announced the list of companies selected to participate in an unprecedented program under which they would receive weapons-grade plutonium recycled from U.S. nuclear weapons for use in nuclear power plants.
The five companies tapped to begin advanced negotiations on plans for transferring and securing the plutonium are Oklo, Standard Nuclear, Exodys Energy, Shine Technologies, and Flibe Energy, the department said June 22 in a written statement. Secretary of Energy Chris Wright was previously on the board at Oklo but divested his shares before joining the Energy Department.
“Surplus plutonium was once viewed solely as a nuclear liability and a multi-billion-dollar financial drain on taxpayers—but it doesn’t have to remain one,” Josh Jarrell, the deputy assistant secretary for the nuclear fuel cycle, said in a statement June 22. “[W]e are redirecting this Cold War legacy to serve as a vital energy asset, powering the next generation of American nuclear innovation.”
In the same press release, the Office of Nuclear Energy stated that the “Surplus Plutonium Utilization Program was … formed to broaden domestic nuclear fuel supplies, spur innovation on American recycling technologies, and unlock private sector funding to fuel America’s nuclear renaissance.”
Although department officials and other advocates say the program will harness the private sector to deliver national energy independence, some members of Congress and nonproliferation experts have expressed concerns that it could create new security risks. “It raises serious weapons proliferation concerns, makes little economic sense, and may adversely affect the nations defense posture,” Sen. Ed Markey (Mass.) and Reps. Don Beyer (Va.) and John Garamendi (Calif.), all Democrats, wrote in a September letter to the department.
In May 2025, President Donald Trump initiated a new nuclear energy policy that aims to promote a “nuclear renaissance” in order to increase domestic energy stability and independence by building new nuclear power reactors.
In October, the Energy Department reported that it had approximately 20 tons of weapons-grade plutonium from dismantled nuclear weapons to make available to U.S. power companies.
As the Energy Department dismantled Cold War-era nuclear weapons over the years, it accumulated more than 50 metric tons of surplus weapons-grade plutonium. The department managed it by blending some stocks into mixed-oxide fuel, and later by diluting and disposing of other stocks.
A May 2025 executive order instructed the energy secretary to stop the dilute-and-dispose program and explore alternatives, including recycling, for surplus plutonium. Another executive order that month called for the Nuclear Regulatory Commission to simplify the licensing process for new reactors.
At that time, Ernest Moniz, a former U.S. energy secretary, warned that allowing companies to use plutonium-based fuels could “Encourage states without nuclear weapons to develop and deploy technologies that are usable in nuclear weapons programs” and “Produce new radioactive waste streams that must be managed, at considerable cost and risk.”
When coupled with pressure on the NRC, this program “could lead to the hasty deployment of advanced reactors with safety and security flaws,” Moniz said. “Additionally, reorganizing and reducing the independence of the NRC could lead to the hasty deployment of advanced reactors with safety and security flaws. A major event would, like those in the past, increase regulatory requirements and set back nuclear energy for a long time.”
In their letter, Markey, Beyer and Garamendi voiced concerns that giving weapons-grade plutonium to private entities would damage the U.S. nonproliferation strategy. Citing policies dating to the Ford and Carter administrations, they highlighted the contradiction in discouraging other states from reprocessing plutonium while doing so itself on domestic soil and noted the potential dangers of rogue states or terrorists accessing this material.
