Anthropic scrambles to contain a significant security breach involving leaked code from its Claude AI agent system.

Anthropic is actively working to contain the fallout from a code leak related to Claude, its flagship AI agent

The leaked code relates to the underlying infrastructure and mechanisms powering Claude's agent capabilities

The incident highlights growing security concerns around proprietary AI systems and intellectual property protection

Details about the leak's scope, origin, and potential impact remain limited pending Anthropic's investigation