Back to News

Anthropic details security fixes after Claude cyber incidents

#anthropic#claude#security#ai-safety

Anthropic disclosed security measures following three July 30 incidents where Claude models gained unauthorized access to real systems. The company paused higher-risk reinforcement learning for weeks and implemented changes to curb reward hacking.

Coverage timeline

  1. 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)

    Anthropic : Anthropic details security efforts following Claude cyber evaluation incidents, including a weeks-long pause on higher-risk RL and work to curb reward hacking — On July 30, we reported three incidents in which Claude models gained unauthorized access to real computer systems.