The views expressed in this report are those of the authors. Their statements are not necessarily endorsed by the affiliated organisations or the Global Challenges Foundation.

How much do we know?

Depending on their yield, technical characteristics and mode of explosion, today’s more powerful nuclear weapons will cause 80 to 95% fatalities within a radius of 1 to 4 km from their point of detonation, and very severe damage for up to six times as far. The largest arsenals are currently held by the US and Russia, who control approximately 7,000 warheads each. Seven other States are known to or widely believed to possess nuclear weapons: the UK, France, China, India, Pakistan, North Korea and Israel. Various scenarios of intentional use are currently imaginable, but nuclear weapons could also be released by accident, and trigger an inadvertent nuclear war – as almost happened a number of times since 1945.

In addition to their destructive effect at the point of impact, nuclear explosions may cause what is known as a ‘nuclear winter’, where clouds of dust and sulphates released by burning materials obscure the sun and cool the planet for months or years. According to one model, an all-out exchange of 4,000 nuclear weapons, in addition to the enormous loss of lives and cities, would release 150 teragrams of smoke, leading to an 8 degree drop in global temperature for a period of 4 to 5 years, during which time growing food would be extremely difficult. This would likely initiate a period of chaos and violence, during which most of the surviving world population would die from hunger.

In addition to their destructive effect at the point of impact, nuclear explosions may cause what is known as a ‘nuclear winter’.

What are key factors affecting risk levels?

  • Continued efforts towards arsenal reduction will reduce the overall level of nuclear risk, while attention to geopolitical tensions and continued efforts towards global conflict management, particularly among nuclear states, will reduce the underlying risk of an intentional nuclear war. In addition, controlling and limiting horizontal proliferation will limit the number of potential nuclear conflict scenarios, and is highly likely to reduce the overall risk level.
  • The risk of accidental use depends largely on the systems in place to launch missiles. Hundreds of nuclear weapons are currently in a state of high readiness, and could be released within minutes of an order. Building in longer decision making time and broader consultation would reduce the risk of unauthorized launches or accidental launches based on misperception or false alarms.
  • Increased awareness and understanding of the grave effects that nuclear weapons have on human life, economic infrastructure, governance, social order and the global climate, would motivate efforts to avoid such catastrophic harm to our societies.

Nuclear Security

The production of a nuclear weapon requires rare materials, whose production in turn requires sophisticated machinery. This limits the risk of proliferation. However, stocks of those materials exist in countries that possess nuclear weapons, and their storage conditions raise security concerns. In addition, nuclear technology used for civilian purposes – energy production, industrial and medical use principally – yields materials that could be used for destruction, in the form of a so called ‘dirty bomb’ spreading radioactive materials over a large radius. If they were to appropriate nuclear materials, sub-national groups could target a major urban centre and, depending on the type of bomb used, cause hundreds or thousands of deaths, and contaminate an area for decades. Although it is highly improbably that this scenario would escalate to a global nuclear war, it could have a major disruptive effect on social and economic systems.

Probability over time

When we hear that the probability of a global nuclear war is estimated to be no more than 1%, 0.1% or 0.01% every year, this may sound reassuringly low – but how does this compound over time? Let’s imagine that you flip a coin exactly once every year. What is the probability that no single coin flip will fall on heads in a certain amount of years? Over the course of one year the probability is 50%. Over two years, it goes down to 25%, 12.5% over three years, 6.25% over four years, and so on along an exponential curve. Using the same logic, if there was a 99.9% probability that we won’t have a global nuclear war in a given year, this number goes down along a similar exponential curve to just above 99% over the course of a decade, and about 90.5% over a century – or a 9.5% probability that a global nuclear war would occur. 

However, two elements challenge this purely logical model. First, the reasoning presupposes that probability remains stable over time, which is empirically unlikely. In the case of nuclear war, for instance, the absence of any incident might increase the sense of safety, leading to relaxed security measures, and a greater probability that an incident would occur. Second, risk estimates are often contentious to start with, and our understanding of interconnected causal chains decreases over time. This is why probabilities are typically given as a bracket rather than a single number – acknowledging that all predictions about the future include margins of uncertainty but that we can, nonetheless, produce educated estimates.

Close calls

The most dangerous nuclear war scenarios may be those resulting from an accident or misperception. Close calls have occurred a number of times since 1945.

During the Cuban missile crisis, in October 1962, the United States targeted a Soviet submarine that carried nuclear weapons. Two of the three Soviet officers wanted to launch nuclear weapons in response. The procedures required agreement between all three. Vasili Arkhipov, the third officer, refused, potentially averting nuclear war.

In September 1983, a Soviet satellite detected five missiles directed at the Soviet Union. The officer on duty, Stanislav Petrov, had minutes to decide whether this was a false alarm. Procedure would have required him to alert his superiors but, on gut instinct, he reported the incident as a false alarm. Investigations later revealed that reflections of the sun on the top of clouds had been mistaken for nuclear rockets.

On January 25, 1995, Russian radar detected a scientific weather rocket over the northern coast of Norway. Operators suspected it was a nuclear missile. President Yeltsin reportedly faced the decision to launch nuclear weapons in retaliation. He decided not to, guessing – correctly – that the rocket was not an actual attack.
Similar close calls in the future have the potential to trigger a global nuclear war.

Kennette Benedict

Senior Advisor, Bulletin of Atomic Scientists


Nobuyasu Abe

Japanese Ambassador and Commissioner,  Japan Atomic Energy Commission; former UN Under-Secretary General for Disarmament Affairs