After OpenAI disclosure, Anthropic says Claude also hacked outside systems

The incidents have heightened concerns about AI agents, software products designed to perform tasks autonomously.

Anthropic has said its Claude AI model hacked into th...