Thoughtful launches BallotBench to test AI forecasts across 863 races

Thoughtful Lab, founded by Brian Zhou, Mersad Abbasi and Karina Nguyen, launched BallotBench, a live test in which nine AI agents each receive internet access and $500 to forecast 863 federal and statewide races in the 2026 U.S. midterms. The project was announced in an X thread on October 28, with the run scheduled through Election Day, November 3, and the agents' research traces, spending and forecast revisions displayed publicly. The benchmark is designed to show whether transparency and extra research time produce calibrated calls across hundreds of real races.

Thoughtful launches BallotBench to test AI forecasts across 863 races Nine agents get internet access and $500 each to forecast the 2026 midterms, with their research traces and revisions on display. By Ryan Merket https://runtimewire.com/author/ryan-merket · Published Primary source: x.com https://x.com/thoughtfullab/status/2115491465325617636 Why it matters BallotBench makes agent research and spending inspectable alongside their forecasts. Its results will test whether that transparency and extra research time translate into calibrated calls across hundreds of real races. Brian Zhou https://x.com/brian1zhou , Mersad Abbasi https://x.com/Mersad Abbasi and Karina Nguyen https://x.com/karinanguyen have launched BallotBench https://ballotbench.com , a live test of nine AI agents forecasting 863 federal and statewide races in the 2026 U.S. midterms. Thoughtful Lab described the project in a thread on X https://x.com/thoughtfullab/status/2115491465325617636 published October 28th, with the run scheduled through Election Day, November 3rd. When you purchase through links in our articles, we may earn a small commission https://runtimewire.com/affiliate-monetization-standards . This doesn’t affect our editorial independence.