January/February 2026
By Samantha Neakrase 

The four biennial Nuclear Security Summits, convening more than 50 heads of government between 2010 and 2016, delivered high-level attention on nuclear terrorism, drove concrete actions to prevent nuclear theft and sabotage, and strengthened the global nuclear security architecture. Yet momentum has waned amid competing priorities, worsened geopolitical dynamics, and weakened multilateral mechanisms. Hopes that new and existing institutions and initiatives would sustain momentum have proven difficult to realize.

Today’s persistent and evolving nuclear security risks—combined with an increasingly complex threat environment—demand renewed attention. Reversing the imbalance between nuclear risks and declining attention will require a more inclusive and flexible strategy that reflects many different priorities, elevates underrepresented voices, and builds new partnerships and coalitions. The 2027 Review Conference of the Amended Convention on the Physical Protection of Nuclear Material (A/CPPNM) and global expansion of nuclear energy offer near-term opportunities to regain focus but will require rethinking traditional approaches.

Summits Drive Progress

The summits drove major progress: Fifteen countries eliminated civilian stocks of weapons-usable nuclear materials, global inventories declined, and many countries strengthened regulations and facility security.¹ The summits also advanced best-practice sharing and peer review, and accelerated treaty ratifications, including by non-summit countries, leading to entry into force of the A/CPPNM, the only international agreement requiring physical protection of nuclear materials and facilities. The summits also produced 11 joint statements (“gift baskets”) in which participants committed to action on key issues such as transport security, minimizing stocks and use of highly enriched uranium (HEU), and insider threats, several of which were later formalized as International Atomic Energy Agency (IAEA) information circulars for wider adoption.²

After the final summit, states hoped that new mechanisms would sustain momentum and address emerging challenges. The summit action plans outlined how international organizations such as the IAEA could carry forward elements of the summits’ agenda. The Nuclear Security Contact Group was created to enable senior officials to continue driving progress without convening heads of government. The new series of IAEA international conferences on nuclear security (ICONS) in 2013, 2016, 2020, and 2024 were intended as regular venues for countries to highlight progress, make commitments, set priorities, and share expertise among practitioners. Following the final 2016 summit push to bring the A/CPPNM into force, countries envisioned periodic review conferences, triggered by the treaty’s entry into force, as a forum to monitor implementation, identify gaps, advance improvements, and address emerging threats.

Declining Attention to Nuclear Security

Although many countries continued improving nuclear security after the summits, the 2020 and 2023 editions of the Nuclear Threat Initiative’s Nuclear Security Index documented a marked decline in progress.³ For example, only two countries have eliminated their weapons-usable nuclear materials since 2016.⁴ Nuclear security has been overshadowed by other global crises and shifting priorities, including climate change and the COVID-19 pandemic. Russia’s 2022 full-scale invasion of Ukraine has further diverted global diplomatic attention and intensified divisions within the IAEA, contributing to increasingly difficult negotiations at recent agency general conferences.

New vehicles for dialogue such as the contact group and the IAEA ICONS showed early promise but have delivered mixed results. The contact group struggled to sustain consistent senior-level engagement and became dormant during the pandemic. The ICONS saw ministerial participation rise in 2020 but decline in 2024, when governments failed—for the first time—to agree on a ministerial declaration.⁵ Although a handful of countries have subscribed to nuclear security information circulars since 2016, most have stagnated.

An Evolving Threat Environment

It is imperative to maintain attention on nuclear security. Even after decades of effort to reduce risks, nuclear terrorism remains one of the greatest international security threats governments face. The risk environment is evolving, and nuclear security, the most effective tool to prevent catastrophe, requires continuous improvement through updated tools and best practices.

Nonstate threats have grown more complex: Beyond large terrorist networks, there is increased risk from self-radicalized individuals, domestic violent extremists, and insiders, who are more difficult to detect.⁶ Some extremists have targeted critical infrastructure, including nuclear facilities.⁷ Expanded access to drones, cyber tools, and advanced technologies such as AI introduce new nuclear security vulnerabilities.

Nuclear security standards and best practices now must also reflect the risks of nuclear facilities in conflict zones.⁸ Russia’s attacks on and occupation of Ukraine’s Zaporizhzhia Nuclear Power Plant violated laws and eroded norms against armed attacks on nuclear facilities.⁹ Attacks on Iranian nuclear facilities further weakened that norm and resulted in degraded nuclear security conditions.¹⁰

The expansion of nuclear energy is positive but poses new security challenges that must be addressed. As more countries pursue nuclear power to meet climate change and energy-security goals, many lack the regulatory frameworks and trained workforce needed to implement robust nuclear security, raising the risk that this expansion will outpace nuclear security preparedness. Small modular reactors and advanced reactors introduce unique security considerations tied to their designs, deployment models, and fuel types.11 Together, these trends underscore the need for renewed global attention on nuclear security at precisely the moment when political focus is waning.

Renewing Momentum

Despite declining high-level attention, there have been some bright spots. The 2022 A/CPPNM review conference adopted a substantive, forward-looking consensus outcome document outlining nuclear security priorities.12 Importantly, a majority of states-parties sent a request to the IAEA to convene a subsequent review conference, signaling support for continued dialogue on implementation of this foundational treaty.13

Although the 2024 ICONS did not produce a consensus ministerial declaration, the co-presidents’ statement gained support from more than 70 countries as reflected in information circular 1233.¹⁴ Two new nuclear security information circulars (1216 and 1217), which recognized the critical nexus between nuclear security and nuclear energy, were each endorsed by 25 or more countries.¹⁵

Three of the original nuclear security information circulars continue to drive engagement and tangible results. Belgium and the United States lead cooperation on insider threats under information circular 908, including at the Second International Symposium on Insider Threat Mitigation in March 2024 with participants from more than 50 countries.¹⁶ The United Kingdom advances transport security cooperation through information circular 909. Norway’s information circular 912 on HEU minimization has supported multiple symposiums, with a fourth planned for March 2026.¹⁷ Together, these efforts show sustained interest and demand for nuclear security cooperation.

Implementation and capacity-building efforts also continue through IAEA advisory services workshops, centers of excellence, and bilateral cooperation programs. Sustained funding for these activities will be essential as nuclear energy expands globally.

Together, these and other initiatives provide a solid foundation for future progress. They show continued willingness to engage constructively, the emergence of new tools and commitments, and sustained work by practitioners to strengthen security. The task ahead is to build on these efforts and bolster them with political and financial support.

Shaping a More Inclusive Narrative

Refocusing global attention on nuclear security requires a larger “nuclear security tent” to create space for more perspectives. Strong nuclear security depends on the engagement of all countries, not just those with weapons-usable nuclear materials or nuclear facilities. Nuclear security has long been framed mainly as a counterterrorism issue driven by those countries, a narrative that remains important but is too narrow, and one that sidelines countries whose priorities center on development, energy security, or peaceful uses.

A more inclusive narrative is needed, one that elevates a range of national and regional perspectives and connects nuclear security to broader priorities. Insights from NTI-hosted dialogues underscore the diversity of viewpoints and the need to respect regional identities. A one-size-fits-all approach is ineffective. Evidence shows that diverse decision-making improves innovation and financial performance.¹⁸ Global nuclear security is no different and will only be as strong as the diversity of countries shaping it. A narrative that reflects many perspectives will be more resilient, politically relevant, and sustainable over time.

Emphasizing Linkages Between Nuclear Security and Climate, Energy, and Development

Growing interest in nuclear power presents an opportunity for a broader nuclear security dialogue. Countries are turning to nuclear energy—including small modular reactors and advanced reactors—to meet net-zero carbon goals, strengthen energy security, and satisfy growing electricity demand. Nuclear technology also advances sustainable development through medical, agricultural, industrial, and infrastructure applications to improve lives around the world.

As more countries seek to adopt nuclear technologies for energy and development, it is critical to emphasize how nuclear security supports, rather than hinders, those goals. Strong nuclear security builds public confidence that nuclear technologies are safe, secure, and not vulnerable to misuse. That confidence is crucial for expanding and sustaining the benefits of peaceful nuclear technology. Concepts such as security by design, which integrate security into all stages of small modular reactor and advanced reactor design and development, will reduce later costs for operators and strengthen nuclear power’s growth. Newcomer countries can reinforce this by signaling to reactor vendors that nuclear security, as well as safeguards and spent fuel management, will be priorities as they review design options.

The link between nuclear power and nuclear security is reflected in recent multilateral statements. The “tripling pledge” approved at the 2023 UN Climate Change Conference (COP28) calls for nuclear expansion to meet “the highest standards of safety, sustainability, security, and non-proliferation” and commits support for countries pursuing new civil nuclear programs under those standards.¹⁹ Together with the two nuclear-energy-related information circulars at the 2024 ICONS, this explicit recognition of the nuclear power-nuclear security nexus provides a strong basis for continued dialogue on why robust security is essential to the successful and sustainable expansion of nuclear power. It reinforces a more inclusive narrative on nuclear security for all countries seeking the benefits of nuclear power.

These linkages should be explored further in international forums focused on climate change, nuclear power, development, and peaceful uses thereof—for example, COP, nuclear power ministerials, the World Economic Forum, and the nuclear Nonproliferation Treaty review process.

Informal and Regional Approaches

Consensus-based multilateralism has limits. In today’s polarized geopolitical environment, consensus alone is insufficient to drive progress, as we saw at recent IAEA general conferences and the 2024 ICONS. This does not mean abandoning consensus where it is achievable but supplementing it with more flexible approaches. Issue-specific coalitions, informal working groups, regional dialogues, and multi-stakeholder partnerships can advance progress when formal negotiations stall. The summit “gift basket” model demonstrated how like-minded countries can achieve outcomes outside consensus. Apart from the two new joint statements at the 2024 ICONS, this model remains underused. Similar approaches may be useful if consensus proves elusive at the 2027 A/CPPNM review conference and the 2028 ICONS.

Although the IAEA will remain central to global nuclear security, regional leadership must play a larger role in addressing the needs of an increasingly diverse nuclear landscape. Regional institutions and coalitions may offer more promising venues for open dialogue, particularly for countries feeling marginalized or disenfranchised in global settings where nuclear security can be perceived as an imposition or barrier to peaceful uses, or is overshadowed by complex political dynamics. Regional engagements can be more relevant to differing regional priorities, which are shaped by common values, history, and priorities. Regional risk assessment hubs could strengthen nuclear security cooperation and analysis. Subregional approaches can account for broader regional divisions around nonproliferation, disarmament, and other security issues.

Sustaining nuclear security over the long term will require strong regional leadership. Regional approaches offer more opportunities for regional leaders to emerge, enabling regions to shape agendas and negotiated outcomes at global meetings such as the 2027 A/CPPNM review conference and the 2028 ICONS. Encouraging greater regional ownership, empowering regional organizations, and ensuring that regional perspectives inform global processes should be core objectives in the coming decade. Platforms such as the African Commission on Nuclear Energy, the ASEAN Network of Regulatory Bodies on Atomic Energy, and the Organization of American States Inter-American Committee Against Terrorism offer promising foundations for these efforts.²⁰

Preparing for the 2027 A/CPPNM Conference

The most significant near-term opportunity to refocus political attention on nuclear security is the 2027 A/CPPNM review conference. As the only legally binding international agreement requiring the physical protection of nuclear materials and facilities, the amended convention remains foundational to the global nuclear security architecture. The 2027 review conference can serve both as a venue to assess progress and as a platform to renew political momentum.²¹

The 2022 A/CPPNM review conference showed that parties can work constructively even in a difficult geopolitical environment. Despite taking place shortly after Russia’s invasion of Ukraine, states-parties adopted a substantive and forward-looking outcome document by consensus.²² Building on that success, the 2027 review conference should aim to: promote the foundational role of the A/CPPNM; build a strong, effective, and sustainable convention regime; strengthen implementation, including Article 14; and encourage new commitments and universalization. In addition, it should strive to be inclusive of diverse perspectives; reaffirm the adequacy of the A/CPPNM; and agree on regular future review conferences as a vehicle for sustained nuclear security dialogue.²³

These objectives should inform the design of the preparatory process. With the 2022 precedent established, there is no need to revisit the rules of procedure or make significant adjustments to the agenda. Preparations can concentrate instead on substantive dialogue and negotiating the outcome document. Regional dialogues in 2026 could help shape priorities and coordinate positions ahead of negotiations.²⁴ States-parties should identify their priorities, consider ways to strengthen the outcome document, and plan how to demonstrate commitment through national statements or new joint statements among countries with shared interests. They should also begin building support for requesting that the IAEA host the next review conference. The most ambitious option would be to institutionalize regular review conferences, with each setting the date for the next one.²⁵

The 2027 review conference will not restore momentum by itself, but it can serve as a powerful anchor, especially if its outcomes feed into the 2028 ICONS. That conference, which is open to all IAEA member states and features more in-depth, cross-sector discussions, complements the review conference. Together, the two events offer critical opportunities to renew political attention.

Sustaining nuclear security will require rethinking global nuclear security dialogues. Regaining momentum amid competing crises and geopolitical tensions demands more flexible, inclusive, and regionally grounded approaches. This means broadening the nuclear security narrative; strengthening links to nuclear power and sustainable development; empowering regional leaders; and recognizing diverse priorities and perspectives. The 2027 A/CPPNM review conference and the 2028 ICONS are key moments to reinforce commitments and reset priorities.

A strong global nuclear security architecture will depend on diverse voices, shared ownership, and sustained attention. Progress is achievable, but only by reframing and adapting nuclear security to a changing world.

ENDNOTES

1. Samantha Neakrase, “The Global Nuclear Security Architecture: Closing Gaps to Build Greater Assurance, Accountability, and Action,” Nuclear Threat Initiative paper, September 2021, p. 1; Matthew Bunn, Nickolas Roth, and William H. Tobey, “Revitalizing Nuclear Security in an Era of Uncertainty,” Belfer Center for Science and International Affairs report, January 2019, pp. 1-4, 15-18, 70.

2. Sara Z. Kutchesfahani, Kelsey Davenport, and Erin Connolly,“The Nuclear Security Summits: An Overview of State Actions to Curb Nuclear Terrorism 2010–2016,” an Arms Control Association and Fissile Materials Working Group report, July 2018.

3. “NTI Nuclear Security Index: Losing Focus in a Disordered World,” Nuclear Threat Initiative, July 2020, p. 8; “NTI Nuclear Security Index: Falling Short in a Dangerous World,” Nuclear Threat Initiative, July 2023, pp. 5-6.

4. Neakrase, Global Nuclear Security Architecture, p. 19; NTI, Nuclear Security Index 2020, p. 39.

5. Information on ministerial attendance at ICONS 2020 and 2024 appears in IAEA General Conference reports: IAEA, “International Conference on Nuclear Security: Sustaining and Strengthening Efforts,” August 31, 2020, p. 1; IAEA, “International Conference on Nuclear Security: Shaping the Future,” p. 1.

6. Rebecca L. Earnhardt, Brendan Hyatt, and Nickolas Roth, “A Threat to Confront: Far-Right Extremists and Nuclear Terrorism,” Bulletin of the Atomic Scientists, January 14, 2021; Sneha Nair, Anna Pluff, and Christina McAllister, “The Threat from Within: An Overview of the Domestic Violent Extremist Threat Facing US Nuclear Security Practitioners,” Stimson Center, November 2, 2023.

7. Dan Sullivan, “National Guard ‘neo-Nazi’ aimed to hit Miami nuclear plant, roommate says,” Tampa Bay Times, June 13, 2017.

8. Nickolas Roth, Ross Matzkin-Bridger, and Jessica Bufford, “Nuclear Facilities in Times of Crisis,” Nuclear Threat Initiative, May 2024, p. 1.

9. Ibid., pp. 1, 5, 9.

10. Theft of nuclear materials was a major concern in the aftermath of the strikes. Reports indicated that Iran had moved its roughly 400 kilograms of HEU before the attacks, raising alarms because weapons-usable nuclear material is more vulnerable during conflict and transport, and a new location might lack the protections of the damaged facilities. Insider threats could also be a risk, as individuals with access to the material could be susceptible to bribes. See Edward Helmore, “JD Vance suggests Iran’s uranium stockpile is still intact despite US strikes,” The Guardian, June 24, 2025.

11. World Institute for Nuclear Security, “Security of Advanced Reactors,” August 2020, pp. 26-28, 32; Laura Holgate, “Sustaining Attention on the Security of Nuclear Materials and Facilities: Challenges and Strategies in a Shifting Global Landscape,” NTI Global Dialogue discussion paper, July 2025, pp. 2, 6.

12. IAEA, “2022 Conference of the Parties to the Amendment to the Convention on the Physical Protection of Nuclear Material,” outcome document, April 2022.

13. IAEA, “Communication Dated 24 March 2022 received from the Permanent Mission of the United States of America concerning a Joint Proposal to the IAEA to convene further CPPNM as amended Review Conferences,” March 25, 2022.

14. IAEA, “Communication dated 23 July 2024 received from the Co-Chairs for the preparatory process of the Ministerial Declaration for the 2024 International Conference on Nuclear Security (ICONS) concerning the Statement by the Co-Presidents of the International Conference on Nuclear Security 2024: Shaping the Future, 20-24 May 2024,” July 24, 2024.

15. IAEA, “Joint Statement on Advanced Nuclear Energy Technologies,” June 4, 2024; IAEA, “Joint Statement on the Role of Nuclear Security in Harnessing the Power of Nuclear Energy,” June 4, 2024.

16. “INFCIRC/908: Insider Threat Mitigation,” Insider Threat Mitigation Hub.

17. IAEA, “The Fourth International Symposium on HEU Minimization organized by the Norwegian Radiation and Nuclear Safety Authority (DSA) (organized in cooperation with the IAEA).”

18. Vivian Hunt, Dennis Layton, and Sara Prince, “Why diversity matters,” McKinsey & Company, January 2015, pp. 1-2.

19. At COP28, countries launch “Declaration to Cripple Nuclear Energy,” U.S. Department of Energy, 2023.

20. For ideas on potential regional forums for dialogue, see Irma Arguello, Trevor Findlay, and Hubert Foy, papers prepared for the November 2019 NTI Global Dialogue.

21. Jonathan Herbach and Samantha Pitts-Kiefer, “More Work to Do: A Pathway for Future Progress on Strengthening Nuclear Security,” Arms Control Today, October 2015.

22. Ibid.

23. Samantha Neakrase, “A Roadmap to the 2027 A/CPPNM Review Conference,” Nuclear Threat Initiative, July 2025.

24. NTI and the IAEA hosted regional dialogues in advance of the 2022 Review Conference; similar NTI-led workshops are now underway.

25. Samantha Neakrase, “Strengthening Nuclear Security With a Sustainable CPPNM Regime,” Arms Control Today, June 2019.

January/February 2026
By Sverre Lodgaard and Alexander Nikitin

During the 60 years between 1960 and 2020, some 69 arms control agreements were negotiated and signed in an attempt to limit or otherwise manage the world’s most destructive weapons. Despite their value, by the late 1970s, some of the agreements and treaties began to be suspended, violated, deconstructed, or parties withdrew from them. That process intensified in the 2010s and 2020s, sparking rumors of the death of arms control. Nevertheless, dozens of agreements are still in force, and although some of the paradigms under which they can be understood have faded, this does not mean the disappearance of arms control as a political phenomenon. In the current and any future security system, arms control is a necessary component.

This article, co-authored by one Western and one Russian expert, analyzes 69 arms control agreements and organizes arms control history into six paradigms, which are deconstructed into basic principles and reconstructed as rising and falling curves reflecting world-order changes.

Arms control is aimed at limiting, prohibiting, or eliminating certain types of weapons, as well as framing the geopolitical, military, and technical capabilities of states aiming at preventing or limiting armed conflict. It has proven to be a relatively systemic political process of interaction among states and alliances in formats of information exchange, talks and negotiations, elaboration, conclusion, verification, and implementation of agreements and treaties.

Paradigm theory supposes that international relations are shaped not only by material factors, but also by clusters of ideas, concepts, and visions affecting politics through social interaction. Mostly, the clusters are shaped by themselves and “live” within the political consciousness and political behavior of national elites and international organizations. Thus, similar to paradigms in science, they are tied to certain political communities and historic conditions and rise, shape, dominate, and dissolve over time. States rarely openly negotiate paradigms in a “rational choice” manner; rather, paradigms result de facto from negotiation practices, manifesting and surviving only in a certain political and international atmosphere, such as while certain world-order models prevail.

The distribution of arms control agreements by decade shows that there were two peaks in achieving new agreements and treaties: in the 1970s (18 new agreements) and in the 1990s (24 new agreements), followed by a decrease to four new agreements in the 2010s and zero in the 2020s.

Breaking the arms control universe into paradigms reveals what has become outmoded, what has been rendered irrelevant under current circumstances, what is useful but in need of updating, and what remains relevant and called for. This process also makes it possible to assess if, and to what extent, the paradigms depend on each other. For instance, although in theory stabilization can be achieved through rearmament, in practice it only has been achieved in a disarmament context.

Despite the difficult geopolitical environment, there are options for pursuing another round of U.S.-Russia strategic arms control negotiations, incorporating lessons learned so far. In the global setting, any arms control approach must address military landscapes that are multipolar and multidomain and characterized by rapid technological turnover.

Attempts to analytically cluster arms control agreements have been undertaken by different experts, most notably by Jozef Goldblat and Jayantha Dhanapala.¹ This article identifies six arms control paradigms based upon sets of interconnected principles that are specified for each paradigm.

The Risk-Reduction Paradigm

This paradigm covers agreements (see Table 1) aimed at lowering the risk of military incident and escalation to war. Agreements in this cluster aim to facilitate instant communication between political and military authorities, define “rules of the road” in military-to-military interactions, and institutionalize military-technical cooperation.

The paradigm is based on three principles beginning with capabilities for instant communication. The first arms control agreement was triggered by the 1962 Cuban missile crisis. The need for instant communication to avoid war by accident or misunderstanding led to the 1963 Hotline Agreement. Another agreement to reduce the risk of accidental nuclear war was signed in 1972, followed by similar agreements between France, the Soviet Union and the United Kingdom a few years later.

Second is the principle of setting rules of the road for military movements and operations. The 1972 Incidents at Sea Agreement between the United States and the Soviet Union was adopted to prevent incidents on and over the high seas.² Navy-to-navy cooperation worked well. In 1975, the two countries agreed on rules of behavior to avoid escalation to nuclear war and, in 1989, rules for that purpose were extended to apply to all armed forces in the Dangerous Military Activities Agreement.

The third principle involved institutionalizing security cooperation. The first agreement of this kind was concluded in 1987, when the Soviet Union and the United States established nuclear risk reduction centers to supplement the hotline and the customary diplomatic channels. In 2000, they agreed to establish a joint data exchange center for data from early warning systems and notification of missile launches, but this center remained only partially operational.

The hotline has been used during military crises such as the Arab-Israeli wars of 1967 and 1973, the 1971 war between India and Pakistan, the Soviet intervention in Afghanistan in 1979, the September 11 terrorist attacks, the U.S. invasion of Iraq in 2003, and in connection with alleged Russian hacking ahead of the 2016 U.S. presidential election.

Among the rules-of-the-road agreements, the Incidents at Sea Agreement stands out as particularly useful even though it does not apply to submerged submarines because of their inherently covert nature. U.S. and Soviet/Russian submarines have collided several times.

The nuclear risk reduction centers of 1987 became the vehicle for all agreed missile notifications between Moscow and Washington, as well as the exchange of information pertaining to arms control agreements. The centers remain operational. Another institutionalized arms control mechanism—the Conflict Prevention Centre of the Conference on Security and Cooperation in Europe, based in Vienna—was established to support confidence- and security-building measures and facilitate conciliation of disputes, but it suffered from an imprecise mandate and, in recent years, from the deterioration of great-power relations.

None of the risk-reduction agreements have been canceled, as of now. Some have been upgraded but others have not, and compliance has been eroding, so there are big question marks about the relevance and utility of many of them under rapidly changing conditions.

The Nonproliferation Paradigm

The nonproliferation paradigm covers bans on the proliferation of weapons of mass destruction (WMDs) and other weapons and military activities to states and nonstate actors, as well as to international territories and the global commons (viz., the high seas, Antarctica, the atmosphere, or outer space).

The paradigm refers to an early group of international agreements, treaties, and conventions, the crowning achievement being the nuclear Nonproliferation Treaty (NPT) signed in 1968 (Table 2). Geographical prohibitions include the spread of military activities to Antarctica and to the moon, nuclear weapons to outer space, and nuclear tests in water, in the atmosphere, or on land. A series of treaties on nuclear-weapon-free zones (NWFZs) later extended this line to cover practically all of the Southern Hemisphere and some parts of the Northern Hemisphere.

This paradigm is also based on three principles, beginning with prohibiting WMD proliferation to the global commons and areas under special international jurisdiction. For instance, the Seabed Treaty of 1971 prohibits the emplacement of WMDs on the ocean floor and in its subsoil. The Outer Space Treaty banned WMD orbiting around Earth, and the Moon Treaty prohibited bringing and locating WMDs on Earth’s satellite. The Antarctic Treaty of 1959 established an international regime to regulate scientific activities, protect the environment, and avert territorial conflicts in the region. The arms control provisions—determining that the region shall be used exclusively for peaceful purposes—derive from those objectives.

The second principle, preventing WMD proliferation to more states, centers on the NPT, which became the backbone of a comprehensive regime of national, regional, and multilateral nonproliferation measures. Agreements and treaties on nuclear-weapon-free zones (NWFZs) are more comprehensive realizations of the same principle.

The third principle, preventing WMD proliferation to nonstate actors, is reflected in the 1980 Convention on the Physical Protection of Nuclear Materials, which went into force in 1987 and seeks to prevent such materials from falling into the hands of terrorist organizations. In 2005, an amendment was adopted, expanding the application of control measures from international transport to include nuclear facilities and nuclear material in peaceful domestic use, storage, and transport.

Two out of four agreements concluded in the first decade of arms control, namely the Seabed and the Outer Space Treaties, were of little or no military significance, but the others left a lasting impact. The 1967 Treaty of Tlatelolco became the forerunner for a series of regional NWFZs, and the NPT became the backbone of an increasingly comprehensive regime of nonproliferation measures.

The number of nonproliferation initiatives grew significantly in the 1990s; 11 agreements were reached between 1992 and 1997. Their significance is corroborated by their contents. Belarus, Kazakhstan, and Ukraine acceded to the NPT as non-nuclear-weapon states; the NPT was extended indefinitely; new NWFZ agreements were concluded for Southeast Asia and Africa; the Comprehensive Test Ban Treaty (CTBT) was signed; and the International Atomic Energy Agency (IAEA) introduced the Model Additional Safeguards Protocol. Finally, South Africa reverted to non-nuclear-weapon state status and joined the NPT on its own initiative.

The Limitation Paradigm

The limitation paradigm covers quantitative ceilings on the future expansion of arsenals and differs from agreements that physically cut quantities or eliminate entire classes of existing weapons. This paradigm includes initial information exchanges on accumulated military capabilities—mostly, but not exclusively, quantitative—and ceilings on the further development and accumulation of assets. The Strategic Arms Limitation Talks and resulting agreements were based on this logic. SALT I limited launchers, whereas SALT II set ceilings primarily on deployed launchers and delivery vehicles, excluding stored and reserve weapons.³

A new principle is the freedom to mix. SALT I allowed the parties to increase their submarine-based forces in return for an equivalent decrease in their land-based ones. All subsequent agreements ending with the 2010 New Strategic Arms Reduction Treaty (New START) have applied variations of this principle.

The most important principle is parity. By freezing different force structures at existing levels, SALT I did not provide for quantitative parity, but SALT II did so, establishing an equal basis for limitations and future reductions.

Finally, this paradigm included the principle of limiting specific military activities. Very importantly, nuclear weapons testing was successively reduced and outlawed by the partial, threshold, and comprehensive test ban treaties.

Limitation was the paradigm of the 1970s. The SALT I limitations lasted for little more than five years. SALT II was tailored to existing expansion plans and easily adhered to but it was never ratified. U.S. President Ronald Reagan declared it void in 1986. The anti-ballistic missile (ABM) limitations eroded in the late 1980s and 1990s and ended in 2002. The limitations on nuclear weapon tests were replaced first by national moratoria and then by the CTBT in 1996.

Together with the ABM Treaty, SALT planted the seeds of strategic arms control. When the Cold War ended, the process was resumed and overtaken by substantial disarmament agreements with intrusive verification provisions. What remains from this category are agreements limiting specific activities: The Nuclear Supplier Group, the Missile Technology Control Regime, the Australia Group, and the Wassenaar Arrangement on export control of conventional arms and dual-use technologies.

The paradigm was short lived for obvious international political reasons. First, increased tensions between East and West in the 1980s brought arms control to a standstill. Then, in the 1990s, the end of the Cold War opened the gates for real disarmament, meaning physical elimination of parts of existing arsenals.

The Stabilization Paradigm

The stabilization paradigm covers agreements to promote strategic stability and crisis prevention by slowing the arms race and promoting disarmament. Early attempts to enhance strategic stability were made in the 1970s when the ABM agreement was concluded. A few years later, another wave of intensified East-West conflict interrupted these efforts, which resumed when the Cold War ended. In a joint statement preceding the Strategic Arms Reduction Treaty I (START I), the Soviet Union and the United States defined the objectives: “To ensure strategic stability, transparency and predictability through further stabilizing reductions … seeking agreements that improve survivability, reduce incentives for a nuclear first strike and implement an appropriate relationship between strategic offenses and defenses.”⁴

This paradigm was based on three principles, starting with preserving strategic stability. Strategic offensive and defensive arms should be configured so that neither side’s defenses could undermine the other side’s retaliatory strike capability. “Preserving mutual deterrence” is shorthand for this concept. The ABM Treaty and the process of de-MIRVing—the reduction of multiple independently targetable reentry vehicles—on ICBMs set in motion by START I were important steps in this direction.

Another principle of this paradigm is reducing incentives for a nuclear-armed state to be the first to use military force of any type in times of crises, otherwise labeled crisis stability. The 1991 Conventional Forces in Europe (CFE) Treaty had a distinct non-offensive profile. The 1987 Intermediate-Range Nuclear Forces (INF) Treaty eliminated an entire class of weapons that were unnerving and destabilizing because of their high accuracy and short flight times.

Reducing incentives to acquire new or more weapons out of fear that the opponent would gain a meaningful advantage, strengthening stability by curbing the arms race, is the third principle. Such incentives were alleviated by arms control and disarmament agreements and in many cases were adjusted by compensatory national measures in order to secure ratification.

When the first arms control agreements were concluded, the parties were not ready to act on any broader understanding of arms control. The idea of joint efforts to enhance the stability of force constellations took more years to mature. It had to be communicated, discussed, and finessed into common understandings. The SALT series of talks began in 1969 not because Soviet and U.S. strategic cultures and political goals were converging, but because arms control made it possible for them to simultaneously promote their irreconcilable interests and their common interests in stabilizing nuclear deterrence to avoid nuclear war. The ABM Treaty, prohibiting nationwide ballistic missile defense, was understood to be a fundamental prerequisite for stable deterrence. In the 1980s, things went from bad to worse when Reagan introduced his Strategic Defense Initiative.⁵

START I brought strategic stability back on target. The number of warheads carried by each strategic missile, known as multiple independently targetable reentry vehicles (MIRVs), was reduced. On the Soviet side, increased mobile ground- and rail-basing was permitted, and deployment of survivable ballistic missile submarines was encouraged on both sides. START II, often referred to as the “de-MIRVing agreement” but never ratified, would have been a major contribution to strategic stability, as well; its most outstanding feature was the reduction of all MIRVed ICBMs.

De-MIRVing thus was set in motion, albeit slowly and halfheartedly. The United States removed the last MIRVs from its land-based missiles in 2014, but kept them on submarine-launched ballistic missiles (SLBMs); survivable submarines are not under the same pressure of early use as land missiles. Russia, meanwhile, maintains limited numbers of MIRVs on both ICBMs and SLBMs.

In the field of conventional forces, the CFE and CFE-1A agreements and subsequent adaptations were another victory for stability and disarmament. The declared objective was to eliminate the disparities in conventional forces that were detrimental to stability. This was achieved by deep and verifiable cuts of offensive weapons.

Mechanisms for the regular exchange of information, verification, and compliance also contributed to confidence-building between former adversaries. But in 2000s, former Warsaw Pact members and former Soviet republics started to join NATO and refused to sign an adapted CFE. Then in 2007, Russia suspended participation in the treaty and, in 2015, ended participation in the treaty’s Joint Consultative Commission.

The Disarmament Paradigm

Different from the principles of limitation, the disarmament paradigm requires the decommissioning, disassembling, and physical destruction of agreed-upon quantities or entire classes of existing weapons and carriers, some of which had been capped at a previous stage. The entire START process from the early 1990s to the 2020s belongs to this category. It cut Russian and U.S. inventories, first from more than 11,000 to about 6,000 accountable warheads, then to 1,550 deployed warheads and 700 delivery vehicles on each side.

The most comprehensive application of the disarmament paradigm may have been the elaboration and adoption of the Chemical Weapons Convention (CWC), requiring full and complete elimination of chemical weapons from the planet. Although the Biological and Toxin Weapons Convention (BTWC) belongs to the same cluster, the lack of verification mechanisms in that treaty significantly reduces the value of that prohibition. Other landmark achievements were the Intermediate-Range Nuclear Forces Treaty, the CFE Treaty, and the presidential nuclear initiatives (see Table 5).

The principles of the disarmament paradigm include the elimination of entire classes of weapons. The 1987 INF Treaty was the first agreement to do this, followed by the CWC in 1993, the Anti-Personnel Landmines Convention in 1997, and the Cluster Munitions Convention in 2008.

The next principle is the elimination of or cuts to delivery vehicles and “accountable” warheads. The START reductions are based on launchers and the means of delivery, establishing counting rules for warheads. The 2002 Strategic Offensive Reductions Treaty (SORT) dealt exclusively with warheads but had no verification provisions and left the reserves untouched.

Another principle is mandating the follow-on destruction of units to be withdrawn. The INF Treaty demanded to destroy all missiles except the guidance systems and the warheads, which were returned to stockpiles for possible reuse. Under START I, decisions on removing weapons from accountability were based on the destruction of launchers or their conversion to carry permitted missiles. New START had similar provisions, but removed missiles need not be destroyed, and bombers could be converted to conventional missions.

Finally, this paradigm included the reduction of reserve arsenals of nondeployed launchers, setting in New START a combined overall limit for deployed and undeployed launchers (800 units, of which up to 700 could be deployed).

Clearly, this was the disarmament decade in arms control history. START I was a big step ahead for nuclear disarmament, in terms of its arms reductions and its comprehensive exchange of information and intrusive verification provisions. The unratified START II aimed to cut the arsenals by another 50 percent from 6,000 “accountable” warheads on each side to between 3,000 and 3,500. START I remained in force until New START succeeded it. Building on START I, New START is known for its reductions, comprehensive information exchange, and intrusive verification provisions that were suspended in times of the Ukrainian crisis.

The 1987 INF agreement eliminated all Soviet and U.S. land-based intermediate-range missiles. Table 5 lists nine disarmament agreements concluded in the 1990s, starting with the comprehensive CFE agreement and including disarmament majors such as the START I, presidential nuclear initiatives, and CWC treaties.

In 2017, a novelty was introduced. Born in protest over the failure of nuclear-weapon states to live up to their disarmament obligations, the treaty for the Prohibition of Nuclear Weapons (TPNW) renounced nuclear deterrence and required that its member states abstain from any involvement in nuclear weapon planning whatsoever. The nuclear-weapon states and their allies opposed this treaty’s creation and ratification and ignored it.

The Confidence-Building Paradigm

The confidence-building paradigm was born during negotiations for the 1975 Helsinki Final Act. The measures were made militarily significant by the Stockholm Conference 10 years later and further enhanced by the Vienna documents. Other important parts of the paradigm are the U.S.-Soviet Nuclear Risk Reduction Centers and the Open Skies Treaty (Table 6).

The risk-reduction and confidence-building agreements may be seen on a continuum stretching from instant crisis management to long-term trust generation. Confidence-building measures were premised on the assumption that no party harbored any intention to change the status quo by military means. This was commonly believed to be the case in Europe in the 1980s but not today. In that sense, confidence-building measures are “good weather” measures, whereas risk-reduction agreements address conflict situations of immediate concern.

The four principles underlying this paradigm start with the exchange of information and observation missions that were required by these agreements and greatly improved the transparency of military deployments and activities.

The next principle promoting predictability, typified by the Stockholm Conference which turned the Helsinki confidence-building measures into an elaborate set of militarily significant measures, widening the scope of information exchange, defining longer prior notification times for large military exercises, and enhancing inspection rights.

Third is the principle of providing security assurances as manifested in the 1994 Budapest Agreement, whereby Russia, the United Kingdom, and the United States extended security assurances to Ukraine and national security assurances to non-nuclear-weapon states summarized in UN Security Council Resolution 984 of 1995.

The final principle is institutionalizing confidence-building measures by establishing structures, such as the Nuclear Risk Reduction Centers (1988), the Vienna Conflict Prevention Center (1991), and the U.S.-Russia Joint Data Exchange Center for early warning systems and notifications of missiles launches (2000).

The Helsinki Final Act, which proclaimed the recognition of existing borders and of different political and economic systems in Europe, established the political framework for confidence-building measures.

When the Stockholm Conference convened, the 1986 arsenal of confidence and security building measures became a watershed achievement: They were, in effect, the beginning of the end of the Cold War. The Vienna Documents of 1990, 1992, 1999, and 2011 later adapted and upgraded these measures. Since then, however, confidence and security building measures have not kept pace with the rapid political and military changes and have lost much of their relevance.

The Erosion of Arms Control

Only eight treaties and conventions were concluded in the 21st century. Among them the SORT (or “Moscow Treaty,” 2002); International Convention for the Suppression of Acts of Terrorism (2005); Cluster Munitions Convention (2008); Central Asia NWFZ (2009); New Start (2010); the Joint Comprehensive Plan of Action (JCPOA) with Iran (2015); and the TPNW (2017). This slowdown is reinforced by the fact that many important agreements have been terminated (see Table 7).

A New Arms Control Paradigm

Arms control paradigms had different fortunes and influence (see Figure 1).

The limitation paradigm is outmoded. In a multipolar world of asymmetrical force postures and multidomain war strategies, parity in select categories of weapons is a straitjacket, launchers are inadequate units of account, and verification by national technical means is not enough to allay suspicions of noncompliance. Moreover, some types of weapons call for radical solutions rather than mere ceilings at existing levels. Sometimes, a total prohibition is easier to verify than reductions.

The conditions for revitalizing the confidence-building paradigm—understood as long-term systemic measures to enhance the transparency and predictability of security relations—are not present either. Globally, such measures do not square easily with big power rivalry and controversial spheres-of-interest policies.

The risk-reduction paradigm tries to come to grips with immediate dangers by instant communication to avoid accidents and by rules of the road to prevent escalation. In a transitioning world replete with serious conflicts, this paradigm has fundamentally important tasks. Assuming that all the big powers want to avoid another world war, a nuclear war in particular, risk-reduction measures are a sine qua non.

The survival of the nonproliferation paradigm rests with the legitimacy of the NPT and the extension of the nonproliferation principle to more states. The five internationally recognized nuclear-weapon states are united in support of the treaty but have failed miserably to live up to their disarmament obligations. The treaty and the international regime connected to it have proven resilient, but betrayed on the disarmament dimension, stuck in the Middle East, and mostly irrelevant to the Asian nuclear-weapons states, the question is whether they will remain intact.⁶

One important nonproliferation track remains stalled: efforts to ensure the nonproliferation of weapons to more environments, especially to outer space. Another track has been underused: More can be done to prevent the proliferation of WMDs and related technologies to nongovernmental actors. Governments tend to be united against nongovernmental competitors. What is required therefore is attentiveness, diligence, and timely initiatives, including overcoming political opposition.

The disarmament paradigm, however, remains relevant. In the current U.S.-Russia dyad, new long-range nuclear and conventional weapons can be fitted under the principles of common ceilings and freedom to mix that were applied under New START. Reduction of reserve arsenals and the destruction of decommissioned units also can be accomplished by proven arms-control methods.

The disarmament and stabilization paradigms are intertwined. Stability was the number one objective of arms control from the beginning. Next was damage limitation should war occur, followed by reduction of the costs of military preparations. Disarmament was desirable but subordinated to stability. However, the history of arms control shows that stability has been achieved by prohibitions (the ABM treaties) and reductions (the START process) and undermined by arms buildup, suggesting that disarmament should be upgraded on the list of arms control objectives.

Thus, agreements that survived and are still valid mostly belong to three of six historic paradigms: risk reduction (10 agreements, all are still alive); nonproliferation (27 agreements, 3 out of order); confidence-building (13 out of 19 agreements are alive). The disarmament paradigm (limitations with real cuts) has been significantly damaged by withdrawals and paralysis(10 out of 19 agreements do not work anymore).

Judging from the history of dying and surviving agreements, it is likely that the next stages of arms control would concentrate on principles of risk reduction, confidence building, and de-escalation techniques, rather than on physical elimination or capping of weapons systems. Separate negotiating tracks may be envisaged for different clusters of weapons. Some tracks could remain bilateral, and some could be multilateral;⁷ some tracks would move slower than others, and some might remain stuck for decades.

Instead of a simplistic request to involve in talks all new carrier systems (U.S. B-21 bombers, Sentinel ICBMs, and Columbia submarines along with Russian Sarmat ICBMs, Oreshnik hypersonic intermediate-range missile, Avantguard boost-glide missile, Kinzhal hypersonic missile, and Poseidon submarines), a relatively new principle could be of system-for-system trade-offs, not numerical ceilings. One side may express readiness not to finalize or not to deploy some of those systems, if the other side would agree to negotiations and make its own comparable (though not necessarily symmetrical) concessions.

The principle of “free mix between categories” or “overall ceiling” was applied within New Start to strategic carriers; each side was allowed to mix freely any quantities of ICBMs, SLBMs and strategic bombers. A new approach could be to apply the free-mix principle to strategic and nonstrategic carriers within an extended overall ceiling. Russia or the United States may want to add to the arsenal new short- or midrange missiles, as they already do. Under such an approach each side could expand their midrange arsenal, in exchange for cutting some of strategic items.

If U.S.-Russia relations improve, then within the contours of future arms control, there could be a return to exchange of data on missile activities. If any country resumes nuclear testing, then an exchange of data on forthcoming or past nuclear tests may become a subject of mutual interest or agreement, either via the Comprehensive Test Ban Treaty Organization, or bilaterally, and lead to revitalizing the data exchange centers established in 1987.

Cyberweapons, artificial intelligence, and new technologies also need to be discussed to identify the risks and advantages they bring.

To reopen space for arms control and disarmament, the world powers must step back from security policies based on zero-sum thinking and worst-case analyses. Above all, there must be a return to cooperative security in some form. To prepare for that possibility, leaders should ensure that the arms control toolbox is updated and prepared for application.

Arms control is grounded in the realization that lasting security is something states must build together with their adversaries or counterparts for mutual gain. To bring it back to life and make proper use of its potential, the world must revert to cooperative security in some form or other. For that to happen, communication across political divides must be encouraged, and leaders must be willing to talk with one another in a businesslike way.

ENDNOTES

1. Jozef Goldblat, Arms Control: A Guide to Negotiations and Agreements, SAGE, 1994, p. 800; Jayantha Dhanapala, “The Hierarchy of Arms Control and Disarmament Treaties,” Denver Journal of International Law & Policy, Vol. 19, No. 1 (2000), p. 5.

2. Later, 11 NATO countries—Canada, France, Germany, Greece, Italy, the Netherlands, Norway, Portugal, Spain, Turkey, and the United Kingdom—concluded bilateral agreements with Russia modeled on the U.S.-Soviet Agreement to Prevent Incidents at Sea.

3. The Strategic Arms Limitation Treaty series of agreements did not limit reserves; such a limitation became a feature in the Strategic Arms Reduction Treaty series of agreements.

4. “Soviet-United States Joint Statement on Future Negotiations on Nuclear and Space Arms and Further Enhancing Strategic Stability,” June 1, 1990.

5. Jeffrey Lewis, “The Nuclear Option: Slowing the New Arms Race Means Compromising on Missile Defenses,” Foreign Affairs, February 22, 2021.

6. See: Joelien Pretorius and Tom Sauer, “Ditch the NPT,” Survival, Vol. 63, No. 4 (2021), pp. 103-24.

7. Ulrich Kühn (ed.), Alexey Arbatov, David Santoro and Tong Zhao, “Trilateral Arms Control? Perspectives from Washington, Moscow, and Beijing,” Institute for Peace Research and Security Policy Research Report #002, 2020.

January/February 2026
By Daryl G. Kimball

For more than five decades, the United States and Russia have observed mutual limits on their deadly long-range nuclear arsenals to mitigate the dangers of nuclear arms racing and nuclear war, but this essential pillar of nuclear strategy is at risk. On Feb. 5, the last remaining treaty limiting the two largest nuclear arsenals—the 2010 New Strategic Arms Reduction Treaty (New START)—will expire. It restricted each side to no more than 1,550 deployed warheads on no more than 700 deployed long-range missiles and bombers.

Seven U.S. presidents from Nixon to Obama concluded major nuclear arms control and nonproliferation agreements. Gerald Ford advanced negotiations on the second Strategic Arms Limitation Treaty and Joe Biden extended New START by five years.

After six years in the White House, however, President Donald Trump has failed to make any progress despite talking often about his desire for “denuclearization” with Russia and China, complaining about the high costs of nuclear weapons, and noting the devastating effects of nuclear conflict.

Worse yet, Trump has raised tensions by issuing nuclear threats, expanding costly plans to modernize and upgrade the U.S. arsenal, suggesting he might resume nuclear explosive testing, and announcing plans for an unworkable national missile defense scheme that will encourage adversaries to build up their offensive nuclear systems.

Nevertheless, in August after his summit with Russian President Vladimir Putin, Trump claimed that “we would like to denuclearize. It’s too much power, and we talked about that.” Questioned about the expiration of New START, Trump said in July, “We are starting to work on that … That is a big problem for the world, when you take off nuclear restrictions.” Yes, it is, but so far, he apparently has not done anything about it.

On Sept. 22, Putin offered a modest way forward, announcing that “Russia is ready to continue to adhere to the central quantitative restrictions under the [New] START Treaty for one year after February 5, 2026” if the United States reciprocates. When asked Oct. 5 about Putin’s proposal, Trump said, “Sounds like a good idea to me.”

Unfortunately, the White House has not yet formally replied to the Kremlin offer. And although the Pentagon recently claimed, incorrectly, that “China has not demonstrated a willingness to advance discussions on nuclear risk reduction measures, bilaterally or multilaterally,” there is no indication that Trump or his team have proposed risk reduction or arms control talks with China.

Trump can still help halt a dangerous arms race that no one can win. But to do so, now is the time to turn his vague “denuclearization” talk into tangible arms control action.

The stakes are high. If Trump fails to respond positively to Russia’s proposal for an interim deal to maintain the New START limits, each side likely will begin increasing the size of its deployed nuclear arsenal for the first time in more than 35 years by uploading additional warheads on existing long-range missiles. Many members of the nuclear-weapons establishment are lobbying for such a buildup.

More nuclear weapons will not make anyone safer. The United States already has a massive, devastating, and largely invulnerable nuclear force that is more than sufficient to deter nuclear attack by China, Russia, and any other nuclear-armed state. Contrary to hype, deploying additional U.S. nuclear weapons would not change Chinese President Xi Jinping’s or Putin’s fundamental deterrence calculus in a future war. Increases in Russian and U.S. strategic forces would further destabilize the mutual balance of nuclear terror; strain the costly, behind-schedule U.S. nuclear modernization program; and push China to accelerate its nuclear buildup.

The continued failure of Beijing, Moscow, and Washington to engage in good-faith arms control and disarmament negotiations also violates their obligations under Article VI of the nuclear Nonproliferation Treaty (NPT) and undermines its long-term viability.

However, if Trump and Putin pledge to maintain mutual restraints on their strategic nuclear arsenals and resume bilateral talks on further nuclear reductions, they could prevent unconstrained competition and provide leverage to press China (as well as France and the United Kingdom) to freeze their forces at the current number of strategic launchers. Undoubtedly, this would bolster the beleaguered NPT regime ahead of the treaty’s 2026 review conference.

Russia and the United States each have fewer than 800 total strategic launchers; China has an estimated 550; and France and the United Kingdom have a combined total of about 100. A mutual freeze on strategic nuclear launchers at these levels would not adversely affect any one country’s ability to deter nuclear attack.

These joint restraint measures would create a more positive environment for talks on further strategic reductions, new restrictions on intermediate-range missiles and tactical nuclear weapons, limits on strategic missile defenses, and other nuclear risk reduction measures, including joint steps to mitigate the risks of AI integration in nuclear command and control.

With the end of New START, Trump and Putin can put the world on a safer path by taking commonsense actions to build down the nuclear danger. 

January/February 2026
By Xiaodon Liang and Daryl G. Kimball  

The United States and Russia have dramatically reduced their nuclear stockpiles since the end of the Cold War, thanks to a series of bilateral arms control agreements that have won the support of Republicans and Democrats alike. But with the expiration of the New Strategic Arms Reduction Treaty (New START) set for Feb. 5, and no bilateral talks on further follow-on agreements to contain or cut U.S. and Russian nuclear arsenals on the horizon, a new era of unconstrained global nuclear competition looms.

After nearly a year of negotiations, New START was signed by presidents Barack Obama and Dmitry Medvedev April 8, 2010. It capped accountable deployed strategic nuclear warheads and bombs at 1,550 and counted each heavy bomber as one warhead. These limits were approximately 30 percent below the 2,200-warhead limit set by the 2002 Strategic Offensive Reductions Treaty (SORT), which did not include any new verification provisions.

New START also limited the two sides to no more than 700 deployed intercontinental ballistic missiles (ICBMs), submarine-launched ballistic missiles (SLBMs), and heavy bombers assigned to nuclear missions. Deployed and nondeployed ICBM launchers, SLBM launchers, and bombers were limited to 800. This additional cap restricted the ability for a “break out” of the treaty by preventing either side from retaining large numbers of non-deployed launchers and bombers.

New START was the first treaty to include provisions for directly counting the number of warheads on a missile. Inspectors were permitted, during short-notice inspections, to require the host country to allow for the counting of re-entry vehicles on an ICBM or SLBM.

The treaty did not constrain missile defense programs although its preamble acknowledges the “interrelationship between strategic offensive arms and strategic defensive arms” and that “current strategic defensive arms do not undermine the viability and effectiveness of the strategic offensive arms of the Parties.”

Given that the 1991 START treaty and its verification provisions expired Dec. 5, 2009, Russia and the United States were interested in re-establishing treaty-mandated verification mechanisms for New START. As General Kevin Chilton, commander of U.S. Strategic Command, testified to Congress in June 2010, “If we don’t get the treaty, [the Russians] are not constrained in their development of force structure and... we have no insight into what they’re doing. So, it’s the worst of both possible worlds.”

Upon signing New START, Obama said: “While the New START treaty is an important first step forward, it is just one step on a longer journey.” It was a prescient statement.

Senate ratification of the treaty was a struggle largely due to opposition from a bloc of senators led by Jon Kyl (R-Ariz.), who demanded a long-term plan and budget for modernizing the U.S. nuclear delivery systems and life-extending warheads. In November 2010, the Obama administration delivered revised estimates for funding National Nuclear Security Administration nuclear weapons complex programs totaling $85 billion for fiscal years 2012-2016, an average of $8.5 billion per year. Today, the NNSA weapons budget consumes $20 billion annually.

On December 22, 2010, following an eight-month-long process and eight days of often intense floor debate, a bipartisan Senate supermajority voted 71-26 in favor of ratification. Kyl still voted "no." The Russian State Duma and Federation Council completed the ratification process Jan. 26, 2011. The treaty entered into force Feb. 5, 2011, and both parties met the treaty’s central limits by the implementation deadline, Feb. 5, 2018.

Although New START modestly cut the U.S. and Russian strategic arsenals, it still left both with extraordinary firepower and further disarmament diplomacy to pursue. As Senator Lamar Alexander (R-Tenn.) noted in December 2010, New START “leaves our country [and Russia] with enough nuclear warheads to blow any attacker to Kingdom Come.”

However, since New START entered into force, Moscow and Washington have misfired on their on-again, off-again discussions on further nuclear reductions. In 2013, Russian President Vladimir Putin rebuffed an Obama initiative to engage in talks designed to achieve a further one-third reduction, and in 2020, U.S. and Russian negotiators resumed wide-ranging arms control talks but failed to reach any agreement.

With the original 10-year lifespan of the treaty about to expire, President Joe Biden and Putin agreed at the 11th hour, Feb. 3, 2021, to exercise the one-time, five-year extension allowed by Article XIV of the treaty. The move came after the first administration of President Donald Trump balked at extending New START with Russia, while seeking to engage China in a three-way arms control negotiation in 2020.

Following Russia's full-scale invasion of Ukraine in 2022, Russia formally notified the United States Feb. 28, 2023, that it would suspend implementation of key New START provisions, including stockpile data exchanges and on-site inspections, which already were suspended temporarily in 2020 by mutual agreement due to the COVID-19 pandemic. Putin said Russia would not resume implementation unless the United States ended support for Ukraine and brought France and the United Kingdom into arms control talks. He later clarified that Russia would continue abiding by the treaty’s central limits.

In January 2025, the U.S. State Department said that, although “Russia [was] probably close to the deployed warhead limit during much of [2024] and may have exceeded the deployed warhead limit by a small number during portions of 2024,” the United States nonetheless “assesses with high confidence that Russia did not engage in any large-scale activity above the Treaty limits.”

With the final New START expiration date now approaching, Putin announced Sept. 22 that Russia is “prepared to continue observing the … central quantitative restrictions” of the treaty for one year after its expiration if the United States “acts in a similar spirit.” The proposal came in the context of talks on ending the conflict in Ukraine and vague statements by Trump that he wanted to maintain limits on nuclear arsenals.

In the absence of such an arrangement, each side will be free to upload additional warheads on existing land- and sea-based missiles, thus increasing their total strategic arsenals for the first time in decades. How the next chapter in the long-history of U.S.-Russian nuclear relations will unfold is yet to be determined.

January/February 2026
By Kelsey Davenport

President Donald Trump said the United States will support new Israeli strikes against Iran’s missile program and threatened to take military action “immediately” if Tehran takes steps to rebuild its nuclear program.

Trump told reporters Dec. 29 that if Iran “will continue with the missiles,” the United States will support Israeli strikes and suggested that Tehran should make a deal with Washington to avoid further attacks.

Trump’s threats came during Israeli Prime Minister Benjamin Netanyahu’s Dec. 29 trip to visit the president at Mar-a-Lago. Before the trip, Israeli officials emphasized that Israel is ready to strike Iran again and suggested that Netanyahu would seek U.S. support for further military action during his visit.

Israel targeted the missile program during its 12-day air campaign against Iran in June, but Tehran was still able to conduct counterstrikes against Israel and the U.S. airbase in Qatar. (See ACT, July/August 2025.)

Since the conflict ended, Iran has announced steps to rebuild and expand its missile program, including equipping its missiles with new technologies. The commander of the Islamic Revolutionary Guard Corps, Major General Mohammad Pakpour, said Dec. 7 that Iran will incorporate stealth technologies into its systems so that they can better evade missile defenses, according to Iran’s state-run Press TV.

Trump’s Dec. 29 comments did not make clear what type of support his administration would provide if Israel strikes Iran’s missile program again, but he suggested that the United States would strike Iran directly if the country took certain steps to reconstitute nuclear activities.

Trump said that he heard that Iran is “trying to build up [its nuclear program] again.” If true, the United States will have “no choice but [to] very quickly eradicate that buildup,” he said.

Although it does not appear that Iran is restarting nuclear activities impacted by the U.S. and Israeli strikes in June, such as uranium enrichment, it is more challenging to assess the status of the country’s nuclear program while Tehran continues to prohibit International Atomic Energy Agency (IAEA) inspectors from accessing bombed sites.

Iran is legally obligated under the nuclear Nonproliferation Treaty (NPT) to implement safeguards on its nuclear program, but it suspended cooperation with the agency following the June attacks. Iran later allowed inspectors to return to sites that were not targeted, such as the Bushehr nuclear power plant, but has yet to provide the agency with reports about its nuclear materials or allow inspectors to access bombed facilities, arguing that safety and security concerns preclude cooperation. (See ACT, October 2025.)

In addition to suspending cooperation, Iran called for the IAEA to revise its safeguards approach to take into account the realities of armed conflict, as the agency continues to press Tehran to allow inspectors to visit nuclear facilities damaged by Israeli and U.S. military strikes.

Behrouz Kamalvandi, a spokesman for the Atomic Energy Organization of Iran (AEOI), said Dec. 9 that after an attack, a country “cannot be expected to immediately allow inspectors into damaged sites, because that could mean handing sensitive information to its enemies.”

Iran has accused the IAEA of providing information used by Israel and the United States in the June strikes but presented no evidence to support those accusations.

IAEA Director-General Rafael Mariano Grossi said in a Dec. 20 interview with RIA Novosti, a Russian state news service, that the agency must assess for itself if Iran’s nuclear sites are unsafe or inaccessible due to the strikes. He reiterated Iran’s legal obligation to allow the IAEA access and said talks with Tehran are ongoing.

In November, the IAEA Board of Governors censured Iran for failing to implement its safeguards obligations and urged Tehran to cooperate with the agency. (See ACT, December 2025.)

Kamalvandi did not provide details about the changes Iran wants the IAEA to make, but he noted that the NPT and its safeguards provisions were drafted for “peaceful conditions,” and said that procedures and processes need to be revised to function under “wartime pressures” when there are unique security interests.

Despite suspending cooperation with the IAEA, Kamalvandi said that Iran remains committed to the NPT and its safeguards agreement.

Esmaeil Baghaei, the Iranian foreign ministry spokesman, reiterated Iran’s commitment to the NPT and said Dec. 22, at his weekly press briefing, that the issues raised by the IAEA require answers from the “parties responsible” for the current situation. Baghaei said that the “perpetrators of illegal and criminal attacks on these facilities” caused the interruption of monitoring.

January/February 2026
By Xiaodon Liang

China will maintain its nuclear capabilities “at the minimum level required for national security” while continuing a policy of no first use of nuclear weapons, a new policy statement claims, even as U.S. intelligence alleges the deployment of intercontinental ballistic missiles (ICBMs) in a third of the silos at China’s new missile bases.

The restatement of Chinese policy came Nov. 27 in an official government white paper on arms control, disarmament, and nonproliferation. The paper is Beijing’s first comprehensive statement on these topics since open-source intelligence analysis identified a large-scale expansion of Chinese nuclear forces in 2021. (See ACT, September 2021).

But the document will disappoint those seeking an official acknowledgement by Beijing of its buildup or an explicit explanation of the logic behind its evolving force posture.

The paper comes close when blaming the United States for “reinforcing its nuclear deterrence and war-fighting capabilities” and “blurring the line between missile defense and strategic offense on purpose,” which constitute “severe threats” to strategic stability.

In response, China is improving “strategic early warning, command and control, missile penetration, and rapid response, as well as [nuclear forces’] survivability,” the document says.

This falls short of confirming a Pentagon claim, reiterated in this year’s edition of an annual report on China’s military, released Dec. 23, that China is making “progress on its attempts to achieve” an “early warning counterstrike” capability “where warning of a missile strike enables a counterstrike launch before an enemy first strike can detonate.”

The report reveals a previously unreported test launch by China of several ICBMs in “quick succession” in December 2024, “indicating the ability to rapidly launch multiple silo-based” missiles. The Pentagon previously noted a similar test in September 2023.

U.S. intelligence continues to assess that China may acquire a force of 1,000 operational nuclear warheads by 2030, the new report says. Although the Chinese white paper does not discuss quantitative requirements, it states that China “never has and never will engage in any nuclear arms race.”

The Pentagon report indicates that China’s strategic forces have deployed DF-31 solid-fueled ICBMs at some 100 silos in new missile fields in the northwest of China. The new missile fields, along with 30 new silos at older fields, add a total of 350 launchers to the land-based leg of China’s strategic arsenal, which also includes around 140 siloed liquid-fueled and road-mobile solid-fueled ICBMs, according to research by the Federation of American Scientists.

The U.S. report estimates that China’s total number of warheads remains in the low 600s, a similar finding to last year. (See ACT, January/February 2025.) U.S. intelligence assesses “a slower rate of [warhead] production when compared to previous years.”

China has renovated and expanded a pit production and warhead assembly site at Pingtong, Sichuan province, over the last few years, the Washington Post reported Dec. 28, citing imagery analysis by the Open Nuclear Network and the Verification Research, Training and Information Center.

The Pentagon report repeats allegations that China is seeking low-yield warheads, for “counterstrikes against military targets” and to “control nuclear escalation” which would enable China to “deter non-nuclear military actions with its nuclear forces.”

The Pentagon also assesses that “Beijing continues to demonstrate no appetite” for arms control talks.

Although the Chinese white paper endorses the “complete prohibition and thorough destruction” of nuclear weapons, it falls back on the traditional Chinese policy that countries possessing the largest nuclear arsenals—Russia and the United States—have a “special and primary” responsibility to reduce their forces first. Beijing has long relied on this formulation to justify avoiding substantive nuclear arms control efforts.

The paper also criticizes extended deterrence and nuclear sharing arrangements, drawing attention once again to China’s proposal for a no-first-use agreement among the five officially recognized nuclear powers (China, France, Russia, the United Kingdom, and the United States).

Of note in the document’s discussion of nonproliferation is the absence of a restatement of China’s support for the goal of denuclearizing the Korean peninsula. Chinese leader Xi Jinping also did not mention denuclearization during a meeting with his North Korean counterpart, Kim Jong Un, in September. (See ACT, October 2025.)

The white paper repeats previous criticism by Chinese officials of the agreement between Australia, the UK, and the United States (AUKUS) to transfer naval nuclear reactors and highly enriched uranium to Australia as part of a nuclear-powered submarine arms sale.

The AUKUS agreement “apparently runs counter to the object and purpose of the [nuclear Nonproliferation Treaty],” the new policy document says.

January/February 2026
By Xiaodon Liang

The Republican-led U.S. Congress approved an annual defense policy bill that green lights $901 billion in discretionary spending and adds more than $2 billion to President Donald Trump’s request for expanded funding for nuclear weapons modernization and strategic missile defense.                                                           

The National Defense Authorization Act for fiscal year 2026, which Trump signed into law Dec. 18, is supplemented by $118 billion in mandatory defense spending approved earlier this year in a budget reconciliation act. It contains various statutory changes relevant to U.S. nuclear forces and their acquisition, while demanding regular briefings on the president’s signature defense program, the “Golden Dome” missile defense system.

Congress will set final budget levels in appropriations bills for fiscal 2026, which remain trapped in negotiations among legislators on a tangle of domestic policy issues.

This year’s defense policy act, which emerged Dec. 7 from negotiations between the House and the Senate, is not as strident as its equivalent from last year in calling for steps to prepare for an arms race. But the legislation reiterates a demand in the fiscal 2025 act for annual briefings on implementation of the recommendations of the 2023 report of the Congressional Commission on the Strategic Posture of the United States, conditioning certain spending on the initiation of these briefings.

The commission report called for planning to upload additional warheads to U.S. intercontinental ballistic missiles (ICBMs), reconvert certain B-52H bombers for a nuclear role, and open missile tubes on Ohio-class ballistic missile submarines, while also acquiring more Long Range Standoff (LRSO) nuclear cruise missiles, B-21 bombers, and Columbia-class ballistic missile submarines, among other proposals to expand the nuclear force. (See ACT, November 2023.)

This year’s authorization act codifies a requirement that the United States indefinitely deploy no fewer than 400 “operationally available” ICBMs, with 150 launch facilities at each of the three current ICBM bases. This goes beyond a restriction that Congress has imposed in previous years barring the use of annually authorized funds to cut the number of ICBMs or reduce their level of readiness.

The act reflects congressional concern regarding the transition from the presently serving Minuteman III ICBM to the future Sentinel ICBM, which has experienced delays and a large cost-estimate increase. (See ACT, September 2024.) Although the Senate’s version of the policy bill had set a target date of 2033 for the Sentinel to reach initial operational capability, the final version of the act drops the deadline. A Nov. 10 article by the defense trade publication Inside Defense also reported that the service is working toward a late 2033 date.

The new ICBM, which the Air Force expects to cost $141 billion in 2020 dollars, will have an annual authorized budget of $5.3 billion, according to the final version of the defense legislation, $1.2 billion more than Trump requested. That total includes funds appropriated earlier this year in a budget reconciliation act. (See ACT, June 2025.)

Given the delay to Sentinel, which the Air Force had indicated would reach initial operational capability by May 2029, the problem of sustaining Minuteman III missiles has garnered increased attention. (See ACT, October 2025.) The act requires the Pentagon to provide an annual report on its strategy for sustaining the Minuteman III until the Sentinel fully replaces the older missile.

The legislation also authorizes an extra $210 million over the president’s budget and sets a 2034 target for the nuclear-capable sea-launched cruise missile to reach initial operational capability, while requiring that the Pentagon and National Nuclear Security Administration (NNSA) provide a “limited number … to meet combatant command requirements” by the end of September 2032.

The Columbia-class ballistic missile submarine program is authorized to spend a total of $11.9 billion across discretionary and reconciliation funding, $710 million more than the president requested.

Although the Senate proposed spending an extra $149 million this year to accelerate the LRSO air-launched nuclear cruise missile program, conference negotiators snubbed the idea. The Senate proposal had included $8 million for advance planning for a conventional variant of the LRSO. Total authorized spending on the missile could reach $1.05 billion in fiscal 2026.

The defense policy act makes several changes to the nuclear weapons enterprise and how it is managed. Statutes governing the Nuclear Weapons Council, a body of five senior Pentagon officials and the administrator of the NNSA, are modified to explicitly entrust the council with the responsibility for “developing options for adjusting the deterrence posture of the United States in response to evolving international security conditions.”

Although the council focused previously on setting priorities for NNSA that reflect Pentagon needs, this year’s statutory changes grant the council more authority to monitor nuclear delivery systems acquisition programs managed by the armed services, as well. The act would grant the council the power to “annually review the plans and budget” of the military departments, whereas previously it only reviewed those of the NNSA.

Last year, Congress modified a senior position within the Pentagon to create a central policy lead for nuclear weapons matters. That role, now titled the assistant secretary of defense for nuclear deterrence, chemical, and biological defense policy and programs, has been further augmented in this year’s defense act by specific instructions to the Pentagon to provide authorities and resources to the new office for overseeing relevant acquisition programs

On Dec. 18, the Senate confirmed Robert Kadlec, the Trump administration’s nominee for the position. Kadlec previously served in the George W. Bush and first Trump administrations as a health and biosecurity expert. As assistant secretary for preparedness and response at the Department of Health and Human Services, Kadlec was an early driver of the Trump administration’s efforts to identify and distribute a vaccine for the COVID-19 pandemic.

Within the NNSA itself, the act creates a rapid capabilities program to “develop new nuclear weapons or modified nuclear weapons that meet military requirements.” According to its new statutory basis, the program will “utilize non-traditional approaches,” adopt “tailored risk-acceptance processes,” “maximize reuse of existing components,” and take other steps to “carry out projects with the goal of achieving first production unit within 5 years of project initiation.”

According to the NNSA’s budget request for fiscal 2026, the agency already has set up rapid capabilities projects under the Stockpile Responsiveness Program “to execute at least two concurrent rapid development activities.”

The act also amends the statutory requirement that NNSA be able to produce 80 plutonium pits per year by specifying that 30 pits be produced at Los Alamos National Laboratories and 50 pits at the future Savannah River Plutonium Processing Facility. NNSA’s plans for pit production are the subject of an environmental impact assessment mandated by the outcome of a recent National Environmental Policy Act lawsuit. (See ACT, November 2024.)

NNSA is now only required to publish a report on the Stockpile Stewardship Management Plan each odd-numbered fiscal year, with Congress eliminating a requirement for shorter updates each even-numbered year.

The Trump administration’s proposal to expand U.S. missile defense capabilities, known as the Golden Dome program, receives wary approval in the defense policy act. The legislation codifies changes to the U.S. policy on missile defense, closely tracking the administration’s Jan. 27 executive order. (See ACT, March 2025.)

The new policy states that the government will “provide for the common defense of the United States and its citizens by deploying and maintaining a next-generation missile defense shield.” The act nullifies previous language that the United States would “rely on nuclear deterrence to address more sophisticated and larger quantity near-peer intercontinental missile threats,” an assurance that attempted to address concerns in Beijing and Moscow that an expanding U.S. missile defense architecture could undermine strategic stability.

Despite this endorsement of the goals of the Golden Dome, the defense committees expressed concern about the program’s viability in demanding that the Pentagon provide an annual report on the system in addition to quarterly briefings.