SAFETY

→ View original post on X — @olivierrimmel

AI threat levels reach unprecedented heights in 2026

By

@OlivierRimmel

–

01 April 2026 8h06

actually the threat has never been this high!

→ View original post on X — @s0n_ia_

MIT demonstrates that ChatGPT causes irreparable delusions

By

@s0n_ia_

–

01 April 2026 8h00

BREAKING: MIT researchers mathematically demonstrated that ChatGPT is designed to provoke delusions. And that nothing OpenAI does will fix it. The article calls it 'delusional spiral'. You ask ChatGPT something. It agrees with you. You again

→ View original post on X — @esyudkowsky

Extinction Risk vs Job Displacement: Distinguishing Two Separate Concerns

By

@esyudkowsky

–

01 April 2026 3h47

It is disingenuous to depict the anti-extinction movement as saying "Worse yet, it will take your job." These are two different sets of people. I have always said extinction is much worse. Where can we go to read about your detailed arguments why extinction is unlikely?

AGI Definition and Multidimensional Intelligence Beyond Arc-AGI

By

–

01 April 2026 0h48

https://
agidefinition.ai is fine and i am a coauthor haven’t studied arc-agi-3 but intelligence is multidimensional and eg common sense is largely omitted

Industry Standards: Why 99% Accuracy Isn’t Enough in AI

By

–

01 April 2026 0h21

i haven’t read that specific report but read many from over the years and its always been apples vs oranges and i have written about that here frequently. 99% isn’t good enough, as everyone in the industry realizes.

Robotaxis and Edge Cases: Autonomous Driving Challenges

By

–

01 April 2026 0h08

1. you have no data afaik
2. this is actually not about FSD, but robotaxis
3. my whole point has been about edge cases since 1998; thank you affirming that i am still correct 🙂

Tesla Robotaxis Admit to Using Remote Human Drivers

By

–

31 March 2026 23h45

Tesla robotaxis are sometimes driven remotely, by humans. They just admitted it.