Planned Obsolescence

Evidence about risk should be transparent

We can’t develop safety standards if we have to rely on opaque judgment

Sep 25 • Ajeya Cotra

103

9

16

August 2026

The Hugging Face attack surprised me

It’s a major warning shot, and might be the last one we get

Aug 28 • Ajeya Cotra

1,035

169

190

Hurtling through 2026

Scoring ten qualitative forecasts five months early

Aug 14 • Ajeya Cotra

56

4

5

July 2026

Total research transparency would be nice

We could learn what's scary and stop doing that

Jul 10 • Ajeya Cotra

43

5

2

June 2026

Could a company overpower nations?

The leading AI company could gain unprecedented power

Jun 5 • Ajeya Cotra

65

14

5

May 2026

Science and speculation

Will we agree about AI risks in time?

May 1 • Ajeya Cotra

67

9

9

April 2026

Six milestones for AI automation

What can AI do on its own, and how well?

Apr 3 • Ajeya Cotra

98

47

13

March 2026

I underestimated AI capabilities (again)

Revisiting a prediction ten months early

Mar 5 • Ajeya Cotra

111

32

9

February 2026

Takeoff speeds rule everything around me

Not all short timelines are created equal

Feb 12

98

12

10

January 2026

AI predictions for 2026

But first, scoring my predictions for 2025

Jan 14 • Ajeya Cotra

68

6

4

Self-sufficient AI

No, we don't "have AGI already." But in any case, we should articulate clearer milestones.

Jan 6 • Ajeya Cotra

67

15

11

November 2024

OpenAI's CBRN tests seem unclear

OpenAI says o1-preview can't meaningfully help novices make chemical and biological weapons. Their test results don’t clearly establish this.

Nov 21, 2024 • Luca Righetti

1