AI ALIGNMENT FORUM
AF

1658
Rachel Freedman
Ω50120
Message
Dialogue
Subscribe

AIS researcher at CHAI. London/Berkeley. Cats are not model-based reinforcement learners.

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by
Newest
No wikitag contributions to display.
ryan_greenblatt's Shortform
Rachel Freedman1y122

It was a secretive program — it wasn’t advertised anywhere, and we had to sign an NDA about its existence (which we have since been released from). I got the impression that this was because OpenAI really wanted to keep the existence of GPT4 under wraps. Anyway, that means I don’t have any proof beyond my word.

Reply
ryan_greenblatt's Shortform
Rachel Freedman1y6-3

OpenAI did this too, with GPT-4 pre-release. It was a small program, though — I think just 5-10 researchers.

Reply
14Your LLM Judge may be biased
2y
0
28CIRL Corrigibility is Fragile
3y
1