The arms control framework that once served as a cornerstone of strategic stability is no longer fit for purpose. With New START (Strategic Arms Reduction Treaty) standing as the last surviving pillar of a Cold War-era arms control architecture, it is clear that the world it was built for no longer exists.
Originally designed to manage bilateral nuclear competition between the United States and the Soviet Union (and later Russia), these agreements are increasingly obsolete in a world shaped by multipolar threats and disruptive technologies. In short, the strategic environment we face today is far more complex than that of the immediate post–Cold War era and a new arms control framework, as I outline below, is needed to confront that reality.
Today’s Arms Control Landscape
As the U.S. Intelligence Community has assessed, China is in the midst of a major nuclear expansion. Current estimates project Beijing will have more than 1,000 operational nuclear warheads by 2030 and as many as 1,500 by 2035. At the same time, advances in space-based systems, artificial intelligence, offensive cyber capabilities, drones, and precision-guided munitions are transforming how nations think about deterrence and conflict. The recent Ukrainian drone strikes on Russian airfields are only the latest reminder of how emerging technologies are changing the strategic equation.
I have long argued that our approach to arms control must evolve in light of these new realities. When I became Assistant Secretary of State for Arms Control in 2014, one of my first priorities was to initiate a broader dialogue on strategic stability and outer space with China. I reorganized the Arms Control Bureau to increase its focus on China, nuclear deterrence, and emerging technologies like space and cyber—domains that are now central to strategic competition.
In 2018, I testified before the House Foreign Affairs Committee on the growing threats posed by Russian and Chinese nuclear modernization. I warned that maintaining a credible U.S. nuclear deterrent was essential to stability. At the same time, as I explained in 2018 and remains the case now, we must also pay attention to the increasing risks posed by anti-satellite weapons, offensive cyber capabilities, and other emerging and disrupting technologies.
Today, I believe the existing arms control regime is effectively dead — and that a new strategic framework is urgently needed. But before we can design that framework, we must ask a fundamental question: What are our objectives, and can arms control help us achieve them?
Objectives for a New Era of Arms Control
From my perspective, any serious arms control framework t must advance four core principles:
- Protect the United States from strategic attack.
- Enhance deterrence.
- Defend our allies and partners from coercion or aggression.
- Prevent miscalculation that could lead to accidental or unintended war.
Unfortunately, the United States has struggled to meet these objectives for the better part of two decades. I say this as someone who held senior roles in both the Obama and Biden administrations. President Obama was, in my view, overly focused on preserving and deepening nuclear reductions — even as the threat environment shifted dramatically. President Trump took the opposite approach, seeking to tear down the existing structure without a credible plan for what should follow.
Given the dramatically changed strategic landscape, I had hoped the Biden administration would undertake a fundamental reassessment of U.S. arms control strategy. Regrettably, that opportunity largely passed without meaningful action.
Where do we go from here? The answer may sound unsatisfying, but we must face reality: meaningful near-term progress on arms control is unlikely. In fact, we may need to “build up” before we can “build down”— reinvesting in our deterrent and our alliances to stabilize the environment before productive arms control can resume. Indeed, like the Soviets during the debate over the deployment of intermediate nuclear forces in Europe in the 1980s, the Russians and Chinese are unlikely to come to the negotiating table until the United States imposes costs on them through the deployment of additional capabilities.
That said, we must begin laying the groundwork for a next-generation arms control regime — one that reflects the challenges of the 2020s, not the 1970s.
Elements of a Forward-Looking Arms Control Framework
Here is what such a framework might include:
- Trilateral, Not Bilateral: The era of bilateral arms control with Russia is over. With China rapidly expanding its nuclear forces, any future agreements must include Beijing. Washington cannot afford to be bound by bilateral limits while China remains unconstrained. Trilateral engagement is essential to avoid a destabilizing imbalance.
- Political Agreements Over Legally Binding Treaties: Given domestic political gridlock and the fast pace of technological change, new legally binding treaties are unlikely. Instead, flexible political agreements — more difficult to enforce, in some cases, but easier to implement and adapt — may offer the best pathway for managing strategic risks and building transparency. Non-binding multilateral arrangements are not unprecedented in the arms control or nuclear non-proliferation domains – the Wassenaar Arrangement, the Helsinki Final Act, the Missile Technology Control Regime (MTCR), and the Joint Comprehensive Plan of Action (prior to U.S. withdrawal) are all examples of successful frameworks that were not binding in character but were of significant strategic importance.
- Prioritize Stability Over Numbers: Arms control in recent decades has focused on reducing warhead counts. While reductions may still be valuable, the primary focus going forward must be stability — minimizing the risk of miscalculation, escalation, or accidental nuclear use. A strategic focus on stability could include, for instance, limiting destabilizing systems like heavy, MIRVed intercontinental ballistic missiles, which were an integral part of the original START I and START II treaties.
- Clarify the Role of Emerging Technologies: Artificial intelligence, cyber capabilities, and space-based systems are not substitutes for nuclear weapons—but they increasingly shape the context in which nuclear decisions are made. During the Cold War, anti-satellite attacks were feared not for their direct impact, but because they could blind states to incoming nuclear strikes. That dynamic still holds. Future frameworks must consider how emerging technologies affect nuclear command, control, and early warning—without necessarily resolving every aspect of their potential use.
- The Golden Dome Dilemma: Can Missile Defense and Arms Control Coexist? Earlier this year, President Trump announced plans to pursue his “Golden Dome” missile defense initiative, directing the Department of Defense to develop a comprehensive next-generation missile shield. The initiative includes expanding current missile defense systems and exploring advanced technologies such as space-based interceptors. Unsurprisingly, Russia and China have voiced strong opposition, viewing the Golden Dome as a direct threat to their strategic deterrents. It is difficult to envision meaningful progress on a future arms control framework that does not address U.S. missile defense in some capacity. That said, if President Trump were to prioritize arms control, he likely has the domestic political capital to negotiate an agreement that includes some form of limits or transparency around missile defenses. For example, the prohibiting or limiting the deployment of space-based missile defense interceptors in exchange for certain concessions from Russia and China. Ultimately, the question is whether he sees striking an arms control deal with Russia and China as central to his legacy.
In short, arms control must evolve from a legacy tool of superpower rivalry to a modern instrument of strategic risk management. That requires intellectual courage, diplomatic creativity, and bipartisan political will — none of which will come easily. But if we are serious about preventing conflict in an era of nuclear multipolarity and technological disruption, we cannot wait for the next crisis to begin building the frameworks we already know we will need.
We must stop clinging to outdated assumptions and confront the uncomfortable reality that the world has changed faster than our institutions. The old logic of bilateral, binding treaties negotiated over decades is ill-suited for a world where nuclear competition is now multipolar, and where technologies like AI and cyber capabilities can upend deterrence in real time.
This is not a call to abandon arms control — far from it. It is a call to reimagine it. To recognize that arms control is not an end in itself, but a means to serve defense and deterrence objectives. To acknowledge that arms control must now account for new actors, new domains, and new technologies.
If we get this wrong — if we continue to treat the post-Cold War framework as sacred — we risk entering a new era of unconstrained arms racing, eroded deterrence, and increased chances of catastrophic miscalculation. But if we get it right, we can forge a new model: one that reflects today’s realities and builds a foundation for peace in a far more dangerous world.
That work must begin now.
