AI Futures Project

Plan A: Suggestions For Further Work

Yesterday we released AI 2040: Plan A, but there’s lots of work left to do.

Jul 10 • Thomas Larsen

50

12

7

AI 2040: Plan A

The least bad plan we currently know of

Jul 9 • Daniel Kokotajlo, Thomas Larsen, Eli Lifland, Romeo Dean, Ryan Greenblatt, and Brendan Halstead

164

45

32

June 2026

A forecast of Chinese DUV and EUV photolithography progress

What we can and can’t learn from ASML.

Jun 19 • Hamish Low, Joshua Turner, and Romeo Dean

31

7

6

April 2026

Q1 2026 Timelines Update

We told you we'd be updating in both directions!

Apr 2 • Daniel Kokotajlo, Eli Lifland, and Brendan Halstead

159

35

13

February 2026

Grading AI 2027’s 2025 Predictions

How has AI progress compared to AI 2027 thus far?

Feb 12 • Eli Lifland and Daniel Kokotajlo

144

29

19

January 2026

Clarifying how our AI timelines forecasts have changed since AI 2027

Correcting common misunderstandings

Jan 27 • Eli Lifland, Daniel Kokotajlo, and Brendan Halstead

64

10

8

Forecast how AI will progress in 2026

Predict benchmark scores, revenues, uplift, and public opinion

Jan 17 • Eli Lifland

31

1

1

What Happens When Superhuman AIs Compete for Control?

....

Jan 11 • Steven Veld

84

44

10

December 2025

AI Futures Model: Dec 2025 Update

We've significantly improved our model of AI timelines and takeoff speeds!

Dec 31, 2025 • Daniel Kokotajlo, Eli Lifland, Brendan Halstead, and Alex Kastner

183

58

17

Early US policy priorities for AGI

Near-term AI policy is confusing, except for these two recommendations

Dec 9, 2025 • Nick Marsh

78

11

10

November 2025

October 2025

Scenario Scrutiny for AI Policy

A call for concrete stress-testing of AI policy proposals

Oct 28, 2025 • Joshua Turner and Daniel Kokotajlo

66

19

6