Punishing language models for their chains of thought?
Oh, you mean for thoughtcrimes?
Brundage jokes about punishing language models for their chains of thought
AISummary
Miles Brundage posts a sarcastic question asking whether anyone is punishing language models for their chains of thought, then calls it thoughtcrime. The post gives no model names, figures, or research details.
Post on XView on X
Miles BrundageVerified on X
@Miles_Brundage
Source: Miles Brundage · x.comPublished · added here