Organisation · nonprofit · US

AI Futures Project

A small research non-profit forecasting AI's trajectory; author of the widely discussed "AI 2027" scenario.

Reference profile

The AI Futures Project is a small research non-profit that forecasts the trajectory of artificial intelligence. It was founded in 2025 by Daniel Kokotajlo, a former governance researcher at OpenAI who resigned in 2024 citing concerns about the pace of development, and it drew wide attention with its first project, "AI 2027" — a detailed narrative scenario, co-authored with Scott Alexander, Eli Lifland, Thomas Larsen and Romeo Dean, tracing a possible path from present systems through automated coding to a late-2027 intelligence explosion, branching into a takeover ending and a more benign one. The group presents the work as a structured forecast rather than a prediction; commentators praised its specificity while others argued its timelines were too aggressive.

Category
Policy, governance & think tanks
Founded
2025
HQ
Berkeley, US