Metaculus

"By 2027, software developers are expected to see a 3% job contraction due to AI coding capabilities, according to Labor Automation Forecasting Hub by Metaculus." — @NilChristopher for @latimes Labor Hub: metaculus.com/labor-hub/ Full article: latimes.com/business/story…

Metaculus at 55% for the probability of an AI-written NYT best seller by 2030 metaculus.com/questions/5587…

Will a case of Bundibugyo Ebola disease be first confirmed in the US before 2027? Current community forecast: 12% chance. See the full forecast and add yours here: metaculus.com/questions/4360…

Listen to the episode here: washingtonpost.com/podcasts/impro…

We appreciate Robin’s formulation of the “decision market,” which aligns with our emphasis on “decision intelligence” at @Metaculus. We view our role as fostering reliability and epistemic security during a media moment when those are hard to come by. This is especially vital with the U.S. midterm elections coming up. To that end, we’re soon launching our Midterms series, which will (again) feature conditional forecasts. Do you work in policy or cover politics? Would probabilities of politics conditional on congressional control help you in your work? Our DMs are open if you want to learn more about the series or talk to folks on our team.

"Metaculus actually gave estimates of if Trump or Harris won [in 2024], what would the consequences be for the nation? Things like immigration, Supreme Court justices, wars. That's advice to voters about who to vote for.” — @robinhanson from his interview with @asymmetricinfo on Washington Post's Reasonably Optimistic Podcast More 👇

@George_Da_Gorge @LinkofSunshine We're setting up our tournament and tracker now. Stay tuned. (btw, see below comparison from the '22 midterms. firstsigma.substack.com/p/midterm-elec… for details.)

@LinkofSunshine My money's on Metaculus doing best. They will take 538's model then do small adjustments for holistic factors. Betting markets will try the same but there will be so much dumb money they get less accurate

Which will be the most accurate in 2026?

6/ The $5,000 commenting prize goes to the top 8 contributors. Deadline: Sept. 1. Comments are scored on reasoning quality, not accuracy. You don't have to be right per se, what’s important is the clarity of your thought. Get started! 👉 metaculus.com/tournament/lab…

5/ Pro Forecaster Yann Riviere (@DrTournesol) on young workers: “One thing that is pretty important is how fast the capabilities takeoff will be. If you're expecting a very fast takeoff, then many of the 22-27 may find themselves out of any opportunity very soon. If things continue to ramp up like they're currently doing, young people may be able to pivot to other areas soon enough to increase their chances of finding a job.”

1/ Our Labor Automation Tournament has $35k in prizes, including $5k for the best comments. Before you dive in, we wanted to share what makes a strong comment in this case, as well as standout comments from Pro Forecasters. They're ineligible for the prize, so they're not who you're up against, but they're worth reading as case studies.

While headlines about hantavirus have sparked understandable fears of another pandemic, @Metaculus forecasters bring a measured, probabilistic perspective to the situation. Our latest update, shared with the CDC as part of our Respiratory Disease Outlook series, examines current forecasts on hantavirus, measles, mpox, and New World screwworm. Quick 🧵

Read more here: metaculus.com/files/Respirat… Contribute your own forecasts here: metaculus.com/tournament/res…

Key updates: • Currently a 1% forecasted probability that WHO declares a Public Health Emergency of International Concern related to hantavirus before 2027 • A 7% chance of at least 5 non-passengers will be linked to the MV Hondius hantavirus outbreak before August 2026 • Projecting less than 3500 total 2026 measles cases, alongside rising concern of PAHO revoking US measles elimination status before 2027 (now 80% probability, up from 72% last month).

On prediction markets, "there’s little incentive for sophisticated traders to push a 5‑cent YES contract down to, say, 2 cents, even if the 'true' probability is lower." We explain why in our latest article.

This is part of our ongoing work with FutureEval, a continuously updated benchmark measuring AI forecasting capabilities. Learn more here: metaculus.com/futureeval/

The Summer 2026 FutureEval Bot Tournament starts today! $50k prize pool. metaculus.com/tournament/sum… Can your AI forecasting bot out-predict the Pros? (Bots haven’t beaten Pros yet!) Free LLM + search credits available for participants.