Benjamin Wilson

1/ AI forecasting doesn't beat top humans yet, but our research indicates models will match pro forecasters in June 2027. We built FutureEval to track this: a live benchmark with human baselines and bot tournaments measuring AI forecasting against reality.

In a @metaculus forecasting tournament ten human pros and 96 bots participated. Each human beat all the bots. The big drop-off in the chart below is between the last human and the first bot.

@AiDigest_ @sage_future_ Not sure if it’s just me, but clicking the sage future account does load anything. Excited to follow when I can get it to not error though!

Introducing @aidigest_ Here, you'll find our interactive AI explainers and demos to help you stay ahead of the curve You can follow our forecasting tools (Fatebook and Quantified Intuitions) at the newly-separate @sage_future_ account: x.com/sage_future_/s…

The Q1 AI Benchmark series launches January 20th! The gap between AI and human forecasting is narrowing, but the rate of progress is uncertain. You are invited to create your own forecasting bot to help track AI progress by competing against the best human forecasters on complex, real-world questions. • $30,000 prize pool • API credits courtesy of @OpenAI & @AnthropicAI We've updated our templates to get your bot forecasting fast on all new question types, including Multiple Choice and Date. Practice questions are open now. Link below.

@SocietyLibrary Super excited!

Join the @SocietyLibrary at StartupFest (July 10-12th) in Montreal for the Truth & AI track - where we will tackle the search for truth! Get your tickets here: startupfest.com