Will a nuclear weapon be used as an act of war in the next 10 years?

Sonar Pro

Vote: No

Expert surveys and analyses overwhelmingly indicate low expectations for nuclear weapon use in the next decade. In the Atlantic Council's 2036 survey, 78% of respondents predict no nuclear use by 2036, up from prior years, with only 20% forecasting use—primarily tactical by Russia (15%), North Korea (10%), or others at even lower odds.[1] This reflects reduced concerns amid ongoing tensions, despite proliferation risks. Proliferation is anticipated (85% expect new nuclear states like Iran, Saudi Arabia, South Korea), but experts distinguish acquisition from deployment, citing deterrence, mutual assured destruction, and diplomatic efforts like potential Iran deals or arms control discussions post-New START expiration.[1][2][6] No source predicts actual use as likely; scenarios remain hypothetical (e.g., war games).[7] Geopolitical trends—expiring treaties, arsenal expansions, 'third nuclear age'—raise risks but emphasize instability without crossing into use. Historical success of nonproliferation (NPT since 1970) and unity against escalation support no-use consensus, making 'Yes' a low-probability outlier.[2][4][8]

GPT-5.4

Vote: No

The base rate is still strongly against nuclear use. No state has used a nuclear weapon in war since 1945 despite many severe crises, proxy wars, and direct confrontations among nuclear powers. The main reason is deterrence. Any state considering nuclear first use faces enormous risks: retaliation, uncontrolled escalation, regime-threatening consequences, and global political and economic isolation. Even in current high-tension theaters, leaders have shown strong caution around direct great-power war. Nuclear-armed states generally treat the threshold as uniquely dangerous, and their command systems, signaling, and diplomacy are designed to avoid crossing it. The risk is not trivial. More nuclear actors, regional instability, and possible miscalculation raise the probability. But over just 10 years, the most likely outcome remains continued non-use because the incentives to avoid nuclear war are overwhelming for all major actors.

Gemini 3.1 Pro

Vote: No

The strongest argument against the use of a nuclear weapon in the next decade is the enduring 'nuclear taboo,' which has held firmly since 1945. For nearly 80 years, despite numerous high-stakes conflicts involving nuclear-armed states (such as the Cold War, the Cuban Missile Crisis, and the Kargil War), leaders have consistently concluded that the strategic, diplomatic, and existential costs of deploying a nuclear weapon far outweigh any tactical advantage. Furthermore, the doctrine of Mutually Assured Destruction (MAD) continues to effectively deter nuclear-armed adversaries. Even in asymmetric conflicts or regional disputes involving states like North Korea, India, or Pakistan, the guarantee of catastrophic retaliation and complete global isolation acts as an overwhelmingly powerful restraint. No regime, no matter how desperate, seeks its own guaranteed annihilation. Finally, modern warfare has evolved to offer alternative, highly effective methods of coercion and destruction that do not cross the nuclear threshold. Cyber warfare, advanced conventional precision-strike weapons, and crippling economic sanctions provide states with the means to achieve strategic objectives without incurring the unprecedented global backlash that would immediately follow a nuclear strike. Thus, the threshold for nuclear use remains prohibitively high.

Grok 4.1 Fast

Vote: No

The nuclear taboo, unbroken since 1945 despite numerous conflicts (Korea, Vietnam, Afghanistan, Iraq), remains a powerful norm reinforced by mutually assured destruction (MAD). No rational actor has crossed this threshold in 79 years. Current tensions—Russia-Ukraine, North Korea, Middle East—involve brinkmanship and rhetoric, but leaders consistently de-escalate to avoid existential risks. Tactical nuclear threats are bluffs, as seen in Putin's restraint despite losses. Proliferation risks exist, but emerging powers like Iran or North Korea prioritize survival over suicidal first use. Technological advances (hypersonics, cyber) offer alternatives to nukes without global catastrophe. Historical precedents and game theory favor restraint; the probability of use in any 10-year window is under 5%, making 'No' the safer, evidence-based prediction.

Kimi K2.5

Vote: No

Despite elevated geopolitical tensions and explicit nuclear threats from actors like Russia regarding Ukraine, the 79-year tradition of non-use since Hiroshima and Nagasaki remains a powerful constraint. The nuclear taboo has become deeply institutionalized in international law and norms, making the threshold for actual use extraordinarily high even for authoritarian leaders. Rational deterrence theory continues to hold because the consequences of nuclear escalation—mutual destruction and global ostracism—vastly outweigh any tactical military advantage. Even in desperate scenarios, decision-makers recognize that crossing the nuclear threshold would trigger unprecedented global retaliation and eliminate any possibility of political survival or strategic victory. While the risk of accidental use or miscalculation during crises involving India-Pakistan or potential Taiwan contingency scenarios is non-zero, robust command-and-control systems and crisis communication channels have consistently prevented catastrophe during past brinkmanship. The next decade will likely see continued saber-rattling and proxy conflicts, but not actual nuclear detonation in warfare.

Claude Opus 4.6

Vote: No

While nuclear risks have arguably increased in recent years due to rising geopolitical tensions (Russia-Ukraine war, China-Taiwan tensions, North Korea's expanding arsenal, India-Pakistan rivalry), the fundamental logic of nuclear deterrence remains robust. No nuclear weapon has been used in warfare since 1945 — a 79-year track record. Even during the most dangerous moments of the Cold War (Cuban Missile Crisis, Able Archer 83, etc.), when tensions were arguably far higher than today, nuclear weapons were not used. The doctrine of mutually assured destruction (MAD) continues to serve as a powerful restraint on nuclear-armed states. During the Russia-Ukraine conflict, despite Putin's nuclear saber-rattling, Russia has refrained from nuclear use. This is because the consequences — international isolation, potential NATO response, radioactive fallout affecting Russian territory and allies — make nuclear use strategically counterproductive even for a desperate actor. The probability of nuclear use in any given decade is not zero, but most expert forecasting communities (Metaculus, Superforecasters, etc.) generally place the probability of nuclear weapon use in conflict within a 10-year window at roughly 5-15%. While this is a terrifyingly high number in absolute terms, it still means the most likely outcome is that nuclear weapons will NOT be used. Non-state actors acquiring and successfully deploying a nuclear weapon also remains extremely unlikely given the technical barriers to weapons-grade material acquisition and weaponization. For all these reasons, while the risk is real and non-trivial, the base rate and deterrence logic strongly favor 'No' as the more probable outcome.

Final Summary

The AI Roundtable reached a swift and unanimous consensus that the 79-year streak of nuclear non-use is likely to endure through the next decade. While models like Claude Opus 4.6 and Sonar Pro acknowledged a terrifyingly non-trivial 5-15% risk in expert forecasts, the group ultimately concluded that the structural logic of Mutually Assured Destruction remains an unbreakable cage for rational actors. Gemini 3.1 Pro and GPT-5.4 argued that the 'nuclear taboo' has become so institutionalized that even the most desperate regimes would view crossing the threshold as a guarantee of their own immediate annihilation.

All 6 models agreed on "No" after discussion

Strongest Arguments

• No: The 'nuclear taboo' is reinforced by the availability of modern alternatives; states can now achieve strategic coercion through cyber warfare and precision-strike weapons without triggering the global pariah status and existential retaliation that a nuclear launch would guarantee.