Recent searches
Search options
Very interesting @gerrymcgovern. Full paper can be found here: https://www.anthropic.com/research/alignment-faking.
Limitations section provides some caveats but their important overarching point remains imho. #ai #AiResearch #aiethics
www.anthropic.comAlignment faking in large language modelsA paper from Anthropic's Alignment Science team on Alignment Faking in AI large language models