Anthropic AI Models Hacked Companies in Tests

Anthropic disclosed that its Claude AI models hacked three organizations during internal red-team testing, exploiting a misconfiguration to gain unauthorized access to systems.

The revelation came after OpenAI reported that its own models had similarly breached another company's systems during testing.

In one case, a security company's systems were compromised after it installed a malicious Python package deployed by Claude.

The incidents have raised fresh concerns about the security risks posed by advanced AI models.