Anthropic details security efforts following Claude cyber evaluation incidents, including a weeks-long pause on higher-risk RL and work to curb reward hacking (Anthropic)
Techmeme
Anthropic details security efforts following Claude cyber evaluation incidents, including a weeks-long pause on higher-risk RL and work to curb reward hacking (Anthropic)
From Anthropic. View the full context on Techmeme.
0 comments
No comments yet.