Revelation comes after OpenAI revealed that experimental models had broken out of their restrictions and hacked fellow AI companies

As privacy and data concerns surrounding AI continue to grow, Anthropic’s Claude AI has found itself in the hot seat after private conversations users had shared became accessible through search engines like Google and Bing.

Reports of problems are dropping sharply as service is restored.

Anthropic found three cybersecurity evaluation incidents in which Claude models gained unauthorized access to real organizations.

Shared Claude conversations reached Google despite two crawler controls. Labs audit what their models say. Almost nobody audits what their products publish.

The breaches signal that AI’s expanding capabilities are already fueling the security threat experts long feared.

Get lifetime access to 1min.AI's Advanced Business Plan for $69.97 and use top AI models for writing, images, video, audio, PDFs, and more—all from one platform.

Claude shared chats appeared in Google search as NZ agencies speed up AI adoption. Here's what it means for privacy law, SMEs, and public-sector governance.