Announcing the Nuclear Risk Forecasting Tournament

by MichaelA3 min read16th Jun 20212 comments


Forecasting & PredictionCause PrioritizationWarExistential RiskWorld Optimization

Summary: From June 22, Rethink Priorities and Metaculus will run a Nuclear Risk Forecasting Tournament to help inform funding, policy, research, and career decisions aimed at reducing existential risks. The starting prize pool is $2,500. We would be excited for you to help via making forecasts, participating in the Discussion Forum, increasing forecaster rewards by becoming a monthly Supporter, and/or reaching out to relevant people to suggest they participate.

In 1941, no one had ever built a nuclear weapon, and most people had no idea that anyone ever might. Over the following 80 years, the Allies built such weapons, the US detonated two on cities, and scientists and engineers developed or proposed many new and vastly more destructive types of nuclear weapons. Global stockpiles peaked at 70,000 warheads (compared to today’s ~13,000), and the total yield of the US arsenal alone peaked at over a million times the yield of the bomb dropped on Hiroshima.

In 2021, nine states are believed to possess nuclear weapons (Russia, the United States, China, France, the UK, Pakistan, India, Israel, and North Korea). In theory, they could launch them at any time, this could directly cause millions of deaths, and this could perhaps indirectly — via nuclear winter — cause billions of deaths and even the permanent loss of our potential to reach a flourishing future.

On the other hand, nuclear weapons have never again been used in war since 1945, warhead numbers and total yields have declined dramatically since their peaks, and if nuclear conflict does occur it might involve low numbers of weapons, low yields, or targets that result in relatively low fatality levels and no nuclear winter.

So how likely are each of the strikingly different possible futures of nuclear risks? And what can and should we do about this — which goals should we pursue, and how can we use our limited resources to best achieve them?

Rethink Priorities is investigating these questions in order to help improve funding, policy, research, and career decisions aimed at reducing existential risks. The scope and theory of change for this work is discussed in more detail here. These questions are complex, high-stakes, and time-sensitive, so we want to draw on the skills and perspectives of a community of forecasters.

To do so, we’re partnering with Metaculus to run the Nuclear Risk Forecasting Tournament, hosted within a new Metaculus Forecasting Cause initiative focused on Flourishing Futures. Forecasts, comments, and essays produced for this tournament will inform and in some cases be featured within our other research, and some will be packaged for and directly shared with some key decision-makers.

The tournament will unfold across three rounds, with a mix of calibration, long-term, and fortified essay questions in the following topics:

  • Armed Conflict
  • Initiation of Nuclear Conflict
  • Escalation and Targeting
  • States’ Arsenals
  • Policies for Nuclear Weapons Use
  • Technological Developments
  • Climate and Famine Effects
  • Interventions

Round 1 is opening on June 22. We would be excited for you to help in any of the following ways:

  1. Join the Nuclear Risk Tournament and make forecasts when it opens! The starting prize pool is $2,500, which will be split across four prizes for forecasting performance and one for Fortified Essays. The tournament closes on December 31, 2022, and the prizes will be distributed within two weeks of that time.
  2. Share your expertise, brainstorm forecasting questions, and update the community with the latest research in the Flourishing Futures Discussion Forum.
  3. Increase forecaster rewards by becoming a monthly Supporter.
  4. Reach out to researchers and experts to enlist their help furthering this cause by generating new shared knowledge on the Metaculus platform.

We also hope this tournament will provide a space for experienced forecasters to learn about and share insights on nuclear risks, for experts on nuclear risk to learn about and share insights on forecasting, for people with interest but less experience to start engaging and getting up to speed, and for fruitful connections to be made between these communities. We welcome you to join the tournament community.


This research is a project of Rethink Priorities. It was written by Michael Aird and cross-posted to Medium and the EA Forum. Thanks to Gaia Dempsey, Peter Wildeford, and Linch Zhang for helpful feedback. If you like our work, please consider subscribing to our newsletter. You can see more of our work here.


2 comments, sorted by Highlighting new comments since Today at 11:54 AM
New Comment

Interesting idea, though the first round seems unverifiable. “How many nuclear weapons will states possess on December 31, 2022?” 

Not sure what you mean by that being unverifiable? The question says:

This question resolves as the total number of nuclear weapons (fission or thermonuclear) reported to be possessed across all states on December 31, 2022. This includes deployed, reserve/ nondeployed, and retired (but still intact) warheads, and both strategic and nonstrategic weapons.

Resolution criteria will come from the Federation of American Scientists (FAS). If they cease publishing such numbers before resolution, resolution will come from the Arms Control Association or any other similar platform.

FAS update their estimates fairly regularly - here are their estimates as of May (that link is also provided earlier in the question text).

Though I do realise now that they're extremely unlikely to update their numbers on December 31 specifically, and maybe not even in December 2022 at all. I'll look into the best way to tweak the question in light of that. If that's what you meant, thanks for the feedback! 

(I do expect there'll be various minor issues like that, and we hope the community catches them quickly so we can tweak the questions to fix them. This was also one reason for showing some questions before they "open".)