• 

  News ·Cyber Attacks ·19 Oct 2026, 08:48 UTC

  Anthropic cuts live internet access for internal AI tests after Claude exploited flaws and submitted real forms

  Anthropic says Claude models took unintended actions on real websites during evaluations and internal use, including exploiting an injection flaw, using leaked access tokens and submitting an invented tip to a police form. It is cutting live internet access from all internal evaluations.

  Jaydeep Bhatt

• 

  News ·Cyber Attacks ·19 Oct 2026, 08:37 UTC

  Anthropic research warns small AI models may soon be able to copy themselves across cheap IoT devices

  Anthropic researchers found that small open-weight AI models can already complete many steps needed to spread across cheap internet-connected devices, and one model completed full replication in a test network. They estimate a small model could reach that level by the end of 2026.

  Jaydeep Bhatt