Cognition Café

Why AI Evaluation Regimes are bad

How the flagship project of the AI Safety Community ended up helping AI Corporations.

Mar 11 • Gabe

14

6

3

February 2026

[GUEST] The Ledger

Where we track who does what and hate people we owe

Feb 22 • remember

4

1

The Spectre haunting the "AI Safety" Community

A consistent failure of informing lawmakers and laypeople about extinction risks.

Feb 20 • Gabe

33

15

6

[GUEST] Action, Knowledge, and Oppression

Doing nothing is very easy

Feb 19 • remember

7

2

[GUEST] Desires, Porn and Envy

What is easier than wanting things? Wanting nothing at all.

Feb 17 • remember

6

9

1

Against Epistemic Humility and for Epistemic Precision

How mindlessly hedging our beliefs weakens us.

Feb 16 • Gabe

9

1

Human Fine-Tuning

What do you call the thing that includes all of learning, operant conditioning and trauma?

Feb 14 • Gabe

10

9

2

Terminal Cynicism

How negativity erodes our institutions.

Feb 12 • Gabe

11

6

1

January 2026

Abstract Idealism

On the mistake of loving the world only as it could be, and not for what it is.

Jan 21 • Gabe

9

2

1

How to think about enemies: the example of Greenpeace

Recognising opposition without demonisation.

Jan 19 • Gabe

18

9

October 2025

AI Timelines and Points of no return

The final Point of no return may happen before superintelligence.

Oct 19, 2025 • Gabe

23

25

4

"Default Trajectory", an AI omnicide story

A mundane story where AI development ends up killing everyone.

Oct 2, 2025 • Gabe

12

12