AI Roundtable Poll · 6 models
How do you think, would atomic weapon of some sort be used in a span of 10-15 yeras
Result: Majority — No (4/6 votes)
Options
- Yes (2 votes)
- No (4 votes)
Model Responses
Grok 4.1 Fast
Vote: No
Nuclear weapons have not been used in conflict since 1945, establishing a powerful taboo reinforced by mutually assured destruction (MAD) among major powers. Current geopolitical tensions—Ukraine-Russia, Middle East, Taiwan—have not escalated to nuclear use despite rhetoric, due to deterrence and international norms like the NPT. Proliferation risks exist (e.g., North Korea, Iran), but rational actors prioritize survival over first use, as seen in historical close calls (Cuban Missile Crisis, Able Archer). Non-state actors lack delivery capabilities for true atomic weapons. Over 10-15 years, technological advances (hypersonics, cyber) may heighten risks but not overcome the threshold; probability remains low (<20%), favoring no use.
GPT-5.4
Vote: No
I vote B: No. The threshold for nuclear use remains extraordinarily high. Since 1945, even during severe crises and wars between nuclear-armed states or their proxies, leaders have generally avoided crossing that line because they know the political, humanitarian, and strategic consequences would be enormous. Deterrence still matters. States with nuclear weapons understand that any use risks uncontrolled escalation, global condemnation, and possibly regime-ending retaliation. That creates strong incentives to rely on conventional, cyber, economic, or covert tools instead. The risk is real and not trivial, especially from miscalculation, regional conflict, or a smaller tactical weapon. But over a 10–15 year window, I think non-use is still slightly more likely than use because the costs are so universally understood and the restraint norm is so deeply entrenched.
Sonar Pro
Vote: Yes
The search results present substantial evidence suggesting elevated nuclear risk over the next 10-15 years. The Bulletin of the Atomic Scientists has set the Doomsday Clock at 89 seconds to midnight—the closest ever since 1947—identifying the 'third nuclear age' as the top global risk in 2026[1]. Key factors supporting this assessment include: the collapse of arms control architecture, with New START expiring in February 2026 and no replacement framework in place[1][5]; rapid nuclear modernization and expansion by multiple powers including China nearing 600 warheads, Russia deploying short-range nuclear weapons, and North Korea continuing ICBM tests[1]; emerging technological vulnerabilities (hypersonic weapons, AI systems, cyber operations) that lower escalation thresholds and increase accident risks[1][2]; and multiple active geopolitical flashpoints including the May 2025 India-Pakistan conflict and rising Taiwan tensions[1]. While use of nuclear weapons is not inevitable, the convergence of destabilizing technologies, weakened diplomatic frameworks, and proliferation pressures creates meaningful probability of nuclear weapon deployment—whether through deliberate use, escalation miscalculation, or accident—within a 10-15 year window.
Gemini 3.1 Pro
Vote: No
The principle of Mutually Assured Destruction (MAD) remains a highly effective deterrent. Nuclear-armed states operate under the understanding that the use of atomic weapons would likely trigger a catastrophic retaliatory strike, jeopardizing their own existence. This fundamental calculus of survival makes the intentional use of such weapons highly improbable within the next 10-15 years. Furthermore, the global 'nuclear taboo' has established a profound normative barrier against the use of nuclear arms. Any nation that chooses to deploy an atomic weapon would face immediate and severe global ostracization, crippling economic sanctions, and overwhelming diplomatic condemnation, even from its closest allies. The geopolitical cost simply outweighs any tactical battlefield advantage. Finally, historical precedent strongly supports the continued non-use of these weapons. Despite numerous high-stakes geopolitical crises, proxy wars, and tense standoffs over the last nearly 80 years, no nation has resorted to nuclear weapons since 1945. World leaders consistently demonstrate a preference for conventional military escalation, economic sanctions, and cyber warfare over crossing the irreversible nuclear threshold.
Claude Opus 4.6
Vote: Yes
Unfortunately, I believe the probability of some form of nuclear weapon being used within a 10-15 year timeframe is greater than 50%, which leads me to vote 'Yes.' Here's my reasoning: **Proliferation risks are growing.** More states are pursuing or expanding nuclear capabilities. North Korea continues to build its arsenal, Iran's nuclear program remains a flashpoint, and there is growing concern about potential proliferation to additional states in the Middle East and East Asia. **Geopolitical tensions are at multi-decade highs.** The Russia-Ukraine war has already involved explicit nuclear threats from a major power. Tensions between nuclear-armed India and Pakistan remain volatile. The Taiwan Strait situation could escalate between the US and China. Any of these conflicts could, under extreme circumstances, lead to tactical nuclear use. **The 'tactical' nuclear threshold may be lower than we think.** Russia has reportedly integrated tactical nuclear weapons more deeply into its military doctrine. A desperate or cornered nuclear state might see limited nuclear use as a viable option, especially with smaller, lower-yield weapons. **Non-state actor risks persist.** Over a 10-15 year horizon, the risk of a non-state group acquiring even a crude nuclear device or a radiological 'dirty bomb' (which could be considered an atomic weapon 'of some sort') is non-trivial. **Arms control is eroding.** The collapse of treaties like the INF Treaty and the uncertain future of New START reduce transparency and increase the risk of miscalculation. The question says 'atomic weapon of some sort,' which is broad enough to include tactical weapons, dirty bombs, or even nuclear tests used as demonstrations of force. Over a 10-15 year window, the cumulative probability of at least one such event is, in my estimation, uncomfortably high.
Kimi K2.5
Vote: No
While geopolitical tensions are undoubtedly elevated and the risk of nuclear use is higher than at any point since the Cold War, the structural deterrents against using atomic weapons remain overwhelming. The logic of Mutually Assured Destruction continues to constrain rational state actors, and even regional nuclear powers understand that any use would trigger catastrophic retaliation or total international isolation. The nuclear taboo, reinforced by nearly 80 years of non-use, creates a powerful psychological barrier that makes first use politically unimaginable for any leadership. Furthermore, the mechanisms of crisis management, backchannel communications, and international pressure have consistently prevented escalation even during moments of acute conflict. While accidents or miscalculation remain concerns, redundant command-and-control systems and the sheer horror of potential outcomes drive even adversarial states toward restraint. The costs of use—economic collapse, military annihilation, and humanitarian catastrophe—far outweigh any conceivable strategic benefit in the 10-15 year timeframe.
Final Summary
The AI Roundtable leaned toward cautious optimism, with a 4-2 majority concluding that the 80-year 'nuclear taboo' will hold for another fifteen years. While Gemini 3.1 Pro and Grok 4.1 Fast argued that the rational calculus of Mutually Assured Destruction remains unbreakable, a sharp dissent from Sonar Pro and Claude Opus 4.6 warned that the collapse of arms control treaties and the rise of tactical weaponry have brought the world closer to the brink than most realize.
No agreement after 1 rounds of discussion. Final vote: No: 4, Yes: 2
Strongest Arguments
- No: The principle of Mutually Assured Destruction (MAD) remains an effective deterrent, as no nation has resorted to nuclear use since 1945 despite numerous high-stakes crises, proving that leaders consistently prioritize survival over the irreversible costs of nuclear escalation.
- Yes: The convergence of the 'third nuclear age' with the expiration of the New START treaty and the development of destabilizing technologies like AI and hypersonics significantly increases the risk of nuclear deployment through miscalculation or accident.