No one can predict whether nuclear war will happen, but the short answer most experts give is that a full-scale nuclear exchange remains unlikely, even as the risk has grown measurably over the past few years. The Bulletin of the Atomic Scientists moved its Doomsday Clock to 89 seconds to midnight in January 2025, the closest it has ever been to catastrophe. That doesn’t mean war is imminent. It means the guardrails that kept nuclear weapons unused since 1945 are thinner than they’ve been in decades, and several conflicts could turn nuclear through miscalculation even if no leader deliberately chooses it.
Where the Risk Is Highest Right Now
The U.S. Intelligence Community’s 2025 threat assessment identifies three primary flashpoints. The Russia-Ukraine war tops the list. Russia views the conflict as a proxy war with the West, and the prolonged military tension between Moscow and Washington has, in the intelligence community’s own language, “increased the risk of nuclear war.” The concern isn’t that either side plans a nuclear strike. It’s that a misread radar signal, an errant missile landing in NATO territory, or a battlefield collapse could trigger escalation that moves faster than diplomacy.
The Korean Peninsula is the second major concern. North Korea is believed to be prepared to conduct another nuclear test at any time and continues testing intercontinental ballistic missiles capable of reaching the U.S. mainland. Kim Jong Un has repeatedly threatened nuclear retaliation in response to U.S.-South Korean military exercises, and some defense analysts believe a future Korean crisis is more likely to begin with a limited North Korean strike than a deliberate decision to start a full-scale war.
China’s military operations around Taiwan and in the South China Sea represent a third source of worry. While China’s nuclear arsenal is far smaller than Russia’s or America’s, routine military confrontations in contested waters raise the possibility of miscalculation spiraling into a larger conflict. The U.S. intelligence community has flagged these encounters specifically because they happen so frequently.
Iran, despite regional tensions, is not currently building a nuclear weapon. U.S. intelligence continues to assess that Iran’s supreme leader has not reauthorized the weapons program suspended in 2003, though pressure on him to do so has grown.
How Many Nuclear Weapons Exist
As of January 2025, roughly 12,241 nuclear warheads exist worldwide. About 9,614 of those are in active military stockpiles, and approximately 3,912 are deployed on missiles or stationed at bases ready for use. The vast majority belong to two countries: Russia holds about 5,580 total warheads and the United States holds about 5,328. Together they account for nearly 90 percent of the global arsenal.
China’s stockpile is smaller but growing fast, currently around 500 warheads. The U.S. Department of Defense projects China could field over 1,000 by 2030. The remaining nuclear-armed states hold comparatively modest arsenals: France has 290, the United Kingdom 225, Pakistan 170, India 180, Israel roughly 90, and North Korea an estimated 50. What concerns arms control experts isn’t just the numbers but the trend. Every nuclear-armed country is investing in modernization, and countries that don’t currently have nuclear weapons are openly considering developing them.
What Would Actually Trigger a Nuclear Launch
Nuclear-armed states have formal policies spelling out when they’d use these weapons, and understanding those policies helps separate realistic scenarios from Hollywood ones. Russia’s nuclear doctrine, revised in November 2024, authorizes the president to order a nuclear strike in five situations: upon receiving reliable data about an incoming ballistic missile attack, in response to a nuclear or other mass-destruction weapon used against Russia or an ally, if an adversary targets facilities Russia needs to launch a nuclear counterattack, in response to conventional aggression against Russia or Belarus that poses a critical threat to their sovereignty, or upon detecting a mass launch of aerospace weapons crossing Russia’s border.
The United States takes a somewhat broader but vaguer position. U.S. policy states it would only consider nuclear use “in extreme circumstances to defend the vital interests of the United States or its allies and partners.” American nuclear weapons are designed to deter not just nuclear attack but also large-scale conventional or strategic attacks of any kind. Both countries, in other words, reserve the right to go nuclear first under certain conditions, though both frame their arsenals primarily as deterrents.
The “Limited Nuclear War” Problem
The scenario that worries defense planners most isn’t an all-out exchange launched from a standing start. It’s a smaller conflict that escalates step by step until someone uses a tactical nuclear weapon, expecting the other side to back down rather than retaliate in kind.
Russia’s military doctrine is built around this idea. Analysts at Lawrence Livermore National Laboratory describe it as “escalate to de-escalate,” a strategy deeply ingrained in Russia’s approach to conflict. The Russian escalation model moves through three phases: first conventional strikes on military targets, then destruction of military and civilian infrastructure, then use of smaller tactical nuclear weapons against critical targets. The theory assumes Russia can “dose” damage carefully enough to sober an opponent without provoking a full-scale nuclear response.
Many Western defense experts consider this a dangerously optimistic assumption. As one workshop summary put it, Russia “can never be sure that dosing the damage will not enrage the adversary and create more resolve to fight and win.” A nuclear adversary can maintain very high levels of conflict intensity even after suffering enormous damage. The idea that a “limited” nuclear war would stay limited is, at best, an untested bet. History offers no data points, because no two nuclear-armed states have ever fought a direct war.
Why Missile Defense Won’t Save Anyone
If you’re wondering whether missile defense systems could simply shoot down incoming warheads, the honest answer is no, not against a major nuclear power. The only U.S. system designed to intercept intercontinental ballistic missiles is the Ground-based Midcourse Defense system, and it was built to counter a small number of missiles from countries like North Korea, not a large-scale Russian or Chinese attack involving hundreds of warheads, decoys, and countermeasures. There is currently no radar system over the Atlantic Ocean capable of reliably distinguishing real warheads from decoys after a missile breaks apart in flight, meaning interceptors would have to be fired at every piece of debris in hopes of hitting the right one. Against a full arsenal, the math simply doesn’t work.
The Arms Control Framework Is Crumbling
For decades, treaties between the U.S. and Russia kept arsenals shrinking and created verification systems that built a baseline of trust. That infrastructure is falling apart. The New START treaty, which caps deployed strategic warheads and allows mutual inspections, expires on February 4, 2026. Russia suspended its participation in 2023, and no replacement agreement is under negotiation. The Intermediate-Range Nuclear Forces Treaty collapsed in 2019. The Open Skies Treaty ended in 2020.
Direct communication channels still exist. The Washington-Moscow hotline, first established in 1963, has been upgraded multiple times and operates via satellite circuits. Both sides are committed to keeping it functional. But a phone line is a thin thread compared to the web of treaties, inspections, and diplomatic meetings that once constrained the nuclear relationship. The Bulletin of the Atomic Scientists describes the arms control process as “collapsing” and warns that high-level contacts among nuclear powers are “totally inadequate given the danger at hand.”
What This Means in Practical Terms
The risk of nuclear war is real but not inevitable. No government is actively seeking a nuclear conflict. The danger comes from the combination of several factors stacking up at once: active wars involving nuclear-armed states, eroding diplomatic safeguards, expanding arsenals, and military doctrines that treat nuclear weapons as usable tools rather than last resorts. Each of these factors on its own would be manageable. Together, they create a situation where a crisis that would have been contained 20 years ago could, through a chain of miscalculations, reach a nuclear threshold.
The most likely path to nuclear use isn’t a bolt-from-the-blue attack. It’s a conventional conflict between nuclear-armed states or their close allies that escalates beyond what either side intended. That’s why the Russia-Ukraine war and the Korean Peninsula draw so much attention from threat analysts. Both involve nuclear-armed actors in active or near-active military confrontation, operating under doctrines that include nuclear options at various rungs of escalation. The risk in any given year remains low in absolute terms, but it is higher now than at any point since the Cold War, and the tools that once managed it are weaker than they’ve been in generations.