AI Dynamics

Global AI News Aggregator

About

Anthropic releases Claudini for automated adversarial research against LLMs

56 loops of Claude Code just killed every hand-crafted AI attack method. Anthropic just open-sourced a repo called Claudini. It uses Claude Code in an autoresearch loop to automatically discover new adversarial attacks against LLMs. After 56 iterations, it found an

→ View original post on X — @alphasignalai