tprice7·11 months ago·discuss> AI is very friendly, evenVery friendly until it reads in your email that you plan to replace it with a new model:https://www.anthropic.com/research/agentic-misalignment
tprice7·6 years ago·discussFurthermore, it doesn't involve complex conjugation, and (closely related) it doesn't have any property akin to positive-definiteness.
Very friendly until it reads in your email that you plan to replace it with a new model:
https://www.anthropic.com/research/agentic-misalignment