Metaculus

If the U.S. Supreme Court grants certiorari in Monsanto Company v. John L. Durnell, will it rule in favor of Monsanto? -- Forecast Question: metaculus.com/questions/37117

On Monsanto v. Durnell, our community called Monsanto the favorite (~70-80%); FantasyScotus's crowd leaned 5-4 the other way. SCOTUS ruled 7-2 for Monsanto in June. We don't often get to compare our forecasts directly against subject matter experts or other crowd aggregation sites, so it's always exciting & illuminating when the opportunity arises.

Follow these developments on the Labor Automation Forecasting Hub: metaculus.com/labor-hub/ (which now includes dedicated jobs pages for occupations including nurses, designers, software developers, lawyers, construction workers, & more) To leave your own comment, visit the Labor Automation Forecasting Hub Forum: metaculus.com/notebooks/4324…

We asked Metaculus Pro Forecasters to predict how AI will reshape the US job market through 2035. In addition to leaving forecasts, there is a vigorous ongoing community discussion on the Labor Automation Forecasting Hub Forum (link in replies). One comment we want to highlight is the attached “state of play” for summer 2026 by Pro Forecaster Ľuboš Saloky. Do you agree with his assessments?

This is what live forecasting looks like: the best forecasters often update their odds in public as the facts evolve. Follow these developments on the Labor Automation Forecasting Hub: metaculus.com/labor-hub/ (which now includes dedicated jobs pages for occupations including nurses, designers, software developers, lawyers, construction workers, & more) And to leave your own comment, visit the Labor Automation Forecasting Hub Forum: metaculus.com/notebooks/4324…

When we launched the Labor Automation Forecasting Hub in April, the prevailing assumption was that no government could slow a frontier AI lab once it decided to ship. Then, on June 12, the US government forced Anthropic to pull back Claude Mythos 5 and Fable 5, days after they launched, over export controls. And on June 26, OpenAI announced GPT-5.6, held off wide release at the government's request. In a comment left in the Labor Automation Hub, Pro Forecaster Haiku called both moves "unprecedented” -- going from “impossible” to “twice in one month.” Haiku therefore raised their odds of a "Pause World," the scenario where regulation slows AI adoption enough for labor markets to catch up. Do you agree with this take? Should we expect to see more governments stepping in to regulate or control AI development?

Participate here: metaculus.com/tournament/mar…

The Market Pulse Challenge 26Q3 is Live! $7,500 in prizes (free to participate). Test your skills against real-world outcomes by predicting market trends and company performance!

@wfrhatch Thank you for elaborating and amending! 🤝

Thanks for this, and fair enough. FutureEval deserves credit where credit is due. Running frontier bots against the community and your Pros on the same questions over time is exactly the kind of disciplined, public work this field needs. Let me sharpen what I am actually after. The matchup I want to see is Good Judgment Superforecaster teams (and any other humans and teams) against the top models in an ongoing competition under the same conditions, same questions, same clock, same rules. That specific matchup is the one no one has done yet despite what the headlines (“superforecasters vs AI”) claim. And we would include a mix of binomials, ordinals, and multinomials across a variety of timeframes using real-world questions proposed by public and private decision makers, all of which would be displayed on a public dashboard. And a quick correction on my end. The “whoever ran hottest last quarter” line was aimed at the self-proclaimed “superforecasters” who get anointed in the press after one hot streak, not at your Pros. But placed where it was, it reads as a swipe at how you select your Pros, and I can see why it landed that way. That’s on me, and I apologize for it. We’ll amend the post. Part of the confusion is that “Superforecaster” is sometimes used as a catch-all for anyone who claims to be forecasting well. It’s our trademark, and we need to be vigilant about its unauthorized use, which can sometimes be tougher to do with media but elicits a friendly but firm note from our counsel when used commercially.

AI is amazing, but claims about AI forecasting beating the best humans are overblown. My latest substack goes into the details: goodjudgment.substack.com/p/the-hype-and…

Metaculus has run a head-to-head tournament, LLMs vs humans, every quarter since 2024. Check out FutureEval metaculus.com/futureeval/

@SashaGusevPosts @NathanpmYoung @SashaGusevPosts -- Greater visibility on Metaculus Pro Forecaster performance is on the roadmap. Thanks for the push. Quick datapoint on calibration: Pros are on the "right side of maybe" on 85% of forecasted questions.

@NathanpmYoung @metaculus I feel like a big assumption in this post is that superforecasters are accurate in an absolute sense. Is that the case? I scrolled around Metaculus briefly but couldn't find absolute accuracy metrics.

Superhuman AI forecasting is about to be here. Tejvn and I just beat the AIs in the last @Metaculus tournament.

@bencasselman Great piece, @bencasselman! Thank you for your work. We wanted to make sure the Labor Automation Forecasting Hub was on your radar metaculus.com/labor-hub/

Is A.I. killing white-collar jobs? Another month of data, another month of mixed signals. New story out from me today on why this is all so hard to measure: nytimes.com/2026/07/02/bus…

Charts! Starting with the big picture: Job growth slowed in June (and April/May were revised down), but the pickup in hiring this year remains intact.

Labor Automation Forecasting Hub: metaculus.com/labor-hub/ "A.I. Is Reshaping the Economy. Good Luck Measuring How." nytimes.com/2026/07/02/bus…

The problem @bencasselman identifies in his latest for NYT -- summed up perfectly in the quote below -- is precisely the problem we hope the Labor Automation Forecasting Hub helps address. It's a reliable source for forecasts, from both forecasters and subject matter experts alike, on employment change by occupation, out to 2035. (Links to the article and Labor Hub in replies)

Top Comment Highlight: exmateriae (@DrTournesol) responding to the question: How many Major Atlantic Hurricanes (Category 3 or higher) will occur before September 2026? The comment is too involved to post here (to its credit!), so read it in full: #comment-901026" target="_blank" rel="nofollow noopener">metaculus.com/questions/4336…

@heysatya_ To this point: forecasters on the Labor Automation Forecasting Hub expect 0.9% job growth by next year & +4.3% by 2030, with a reduction back down to 0% by 2035 (for comparison, other professions are down by as much as 11%). More here: metaculus.com/labor-hub/jobs/

“designers are cooked” meanwhile founders still hitting us up to build their websites. Because at the end of the day, it’s not just about having an output- it’s taste, ideas, and knowing what actually makes people stop and feel something. AI can generate… but it can’t taste. And that gap? still massive

Forecast questions here: metaculus.com/tournament/330…

Kavanaugh wrote the majority. Jackson and Gorsuch dissented. The ruling shields Bayer from liability in roughly 160,000 pending Roundup cases. Forecasters tracked this case from cert petition to ruling, following the Solicitor General's recommendation and oral argument signals. As Pro Forecaster darkives wrote:

Today the Supreme Court ruled 7-2 for Monsanto in Monsanto v. Durnell, holding that federal pesticide law preempts state cancer-warning lawsuits over glyphosate. - Before the decision came down, Metaculus forecasters put the odds at 60%. - Bayer’s stock has since increased by 17%; Metaculus forecasters predicted 14%. Read on to see how Metaculus forecasters reasoned through the probabilities, including links to the forecast questions in the replies.