Why Nuclear Deterrence in Asia Is Collapsing
By Patrick M. Cronin, Asia-Pacific Security Chair at the Hudson Institute, Scholar in Residence at Carnegie Mellon University’s Institute for Strategy and Technology (CMIST), and APP Member.
First Published March 31, 2026 on The National Interest.
As the 2027 “Davidson window” approaches, Washington appears less concerned about deterrence failure in Asia than at any point in recent years. The phrase, drawn from former US Indo-Pacific commander Admiral Philip Davidson’s warning about China’s timeline for achieving combat readiness for a Taiwan contingency, once became shorthand for fears of imminent conflict, especially in the wake of Russia’s invasion of Ukraine.
Yet “deterrence” is scarcely mentioned in the US Intelligence Community’s 2026 Annual Threat Assessment, the intelligence community’s capstone document. While it notes that Beijing’s ambitions over Taiwan remain undiminished, it asserts that “Chinese leaders do not currently plan to execute an invasion of Taiwan in 2027, nor do they have a fixed timeline for achieving unification.” The intelligence community appears to have embraced a view of Xi Jinping’s strategy as gradual, positional, and political, aimed at tightening control over what Beijing calls a rogue province, even though the People’s Republic of China has never governed Taiwan.
Admiral Davidson’s warning was never a prediction of impending war. It was a caution about the convergence of two trends: China’s military modernization milestone in 2027 and a potentially declining US force posture, which could create a gap in deterrence. Ironically, on the eve of the People’s Liberation Army’s centenary in 2027, Washington risks becoming overly sanguine about the durability of deterrence in the Indo-Pacific.
Although the Annual Threat Assessment reflects reasonable analytical caution, it risks strategic complacency.
Across both South and East Asia, structural trends are eroding deterrence. A fracturing international order, the expansion of conflict below the nuclear threshold, and the accelerating impact of emerging technologies are together undermining stability. This is no time to slacken efforts to shore up deterrence.
The Failure of Nuclear Deterrence
Recent conflicts suggest the stark lesson that while nuclear weapons may still deter large-scale war, they do not prevent conflict. Indeed, nuclear-armed states appear increasingly willing to incur risk by operating below the nuclear threshold. Russia’s invasion of Ukraine remains the most obvious case. Had Kyiv retained its nuclear arsenal, it is difficult to imagine Russia prosecuting a prolonged war of aggression on its territory.
Iran provides a more recent example of deterrence failure. Tehran exposed its capabilities through premature use, delegated credibility to unreliable proxies, and lingered at a nuclear threshold sufficient to provoke an attack but insufficient to prevent it. What should have imposed caution instead created opportunity for its adversaries. Deterrence did not fail quietly; it collapsed visibly.
More broadly, deterrence is no longer reliably governing the space below major war. This was evident in South Asia in spring 2025.
After a terrorist attack in Pahalgam killed 26 civilians, India launched Operation Sindoor, striking deep inside Pakistan, including near sensitive military infrastructure. The episode demonstrated a higher tolerance for risk on both sides. It reinforced the parlous conclusion that even nuclear-armed rivals believe there is still room to engage in conflict without triggering nuclear escalation. Tensions continue to simmer on the Durand Line between Pakistan and Afghanistan, and last year was not the last time terrorism will prompt military reprisal. India’s subsequent shift toward compellence, outlined in Defence Forces Vision 2047, underscores this trend.
Across recent conflicts, from Ukraine to Iran, states are testing the limits of escalation, probing how far they can go without crossing nuclear red lines. Indeed, a Carnegie Endowment study of nuclear threats in recent years suggests that nuclear-armed governments have grown more adept at manipulating fears of nuclear exchange to pursue conventional aims. Some may think that even a tactical nuclear use might not elicit more than conventional retaliation.
That assumption is profoundly misplaced. As the line between coercion and conflict blurs, and as conventional and nuclear domains become increasingly entangled, ambiguity is rising. And with ambiguity comes a heightened risk of miscalculation. These pressures are global, but nowhere are they more dangerous than in Northeast Asia.
Northeast Asia’s Nuclear Paradox
North Korea has moved beyond deterrence by retaliation toward a doctrine that emphasizes early, limited use to control escalation. Its focus on tactical nuclear weapons, preemption, and potential pre-delegation lowers the threshold for nuclear employment. As Supreme Leader Kim Jong-un made clear in a speech amid the Iran conflict, North Korea’s “nuclear shield…drives the development of all sectors of the country” and guarantees “the dignity of the state, national interests, and ultimate victory.”
Russia acts as an accelerant. Its deepening alignment and“invincible alliance” with North Korea, its reliance on nuclear coercion, and its reported support for Pyongyang’s missile programs are contributing to a more interconnected and volatile deterrence environment. North Korean systems tested in Ukraine are already demonstrating improved survivability and maneuverability, as the KN-23 short-range ballistic missiles are increasingly able to evade interception.
China, meanwhile, is undertaking a rapid and opaque nuclear expansion. Whether Beijing seeks parity or something more ambitious, its trajectory, combined with possible movement toward a launch-on-warning posture, introduces new instability into crisis dynamics.
These trends are not isolated. They reinforce one another, creating an interlinked nuclear system that is more complex, less predictable, and harder to stabilize.
At the strategic level, deterrence still holds. The catastrophic costs of nuclear war continue to constrain major powers. But at the operational level, where crises unfold and decisions are made, deterrence is fragmenting. The central paradox of the current moment is that, as destructive capabilities grow, predictability declines.
The US’ Nuclear Credibility Gap
Credibility, the currency of deterrence, is also steadily fading. Allies increasingly question the reliability of US commitments, even as technological advances fail to deliver greater security. US force posture adjustments have compounded these concerns.
Taiwanese officials have warned that the expenditure of long-range strike systems such as Joint Air-to-Surface Standoff Missiles (JASSMs) and Tomahawks erodes deterrence across the Taiwan Strait. South Korean officials have voiced similar unease over the redeployment of missile defense assets, including Terminal High Altitude Area Defense (THAAD) and Patriot systems, especially given the domestic political and economic costs Seoul previously incurred to host them.
At the same time, the credibility of US extended deterrence, the foundation of security for Japan and South Korea, is under growing strain. The longstanding question of whether the United States would risk Los Angeles to defend Seoul or Tokyo is no longer theoretical, but increasingly being asked in allied capitals.
South Korea is hedging by investing in nuclear-powered submarines, expanding its latent nuclear capability while remaining faithful to the nonproliferation regime. Tokyo, constrained by domestic politics, is strengthening its conventional strike capabilities while reassessing old assumptions, including the role of US nuclear weapons on its territory.
Deterrence rests not only on capability but also on trust and belief. And trust and belief are becoming harder to sustain.
Along with a decline in credibility is the questionable assumption that conflicts in Asia can be geographically contained. The Korean Peninsula, Taiwan Strait, and South China Sea are increasingly interconnected. A Taiwan contingency, for example, would draw US and allied assets away from Korea, weakening deterrence on the peninsula and potentially inviting opportunistic action. In a multi-front scenario, even strong alliances can be stretched thin.
Emerging technologies are exacerbating these risks. Artificial intelligence and autonomous systems are compressing decision-making timelines, reducing the space for human judgment. What once unfolded over days may now occur in minutes. Advances in surveillance are making military movements more visible, increasing incentives for preemption. The logic of “use it or lose it” is becoming more salient. Even the undersea domain, long the foundation of secure second-strike capability, may become more transparent, undermining a key pillar of strategic stability.
Crises are moving faster than the institutions designed to manage them.
How to Strengthen US Nuclear Deterrence Without Breaking It
What, then, is to be done? There is broad agreement on the need to strengthen deterrence, but less consensus on how to do so. Some advocate deeper allied integration. For instance, tighter trilateral coordination among the United States, Japan, and South Korea, enhanced missile defense cooperation, and a more unified defense of the First Island Chain are all prevailing lines of effort for bolstering deterrence in East Asia.
Others emphasize preparing for limited nuclear use, ensuring that alliances can absorb and respond without uncontrolled escalation. But there is also a growing recognition that overreliance on preemption and rapid escalation could sow crisis instability. Efforts to strengthen deterrence may inadvertently accelerate the very dynamics that make it fragile. The policy challenge is fortifying deterrence without breaking it.
Defense modernization programs that envision a “spine” of AI-enabled technologies across all domains, with a focus on resilience of the system, could also invite foes to paralyze the spine, or at least lead them to believe that they might be capable of doing so. Deterrence is no longer a slow-moving, bilateral system. It is a fast, interconnected, multi-actor environment shaped by nuclear modernization, technological disruption, and shifting political commitments.
The greatest danger is not that deterrence collapses outright, but that it fails in ways we do not anticipate. A misinterpreted signal. A limited strike. A decision made too quickly.
The Davidson window may never have been a countdown to war. But it was a warning about vulnerability, and that vulnerability has deepened rather than disappeared. The danger is not that deterrence collapses. It is that it fails in ways we fail to anticipate.
