IBM’s ‘Bob’ AI could be manipulated to download and execute malware



  • IBM’s GenAI tool “Bob” vulnerable to rapid injection indirect attacks in beta testing
  • CLI faces rapid injection risks; IDE exposed to AI-specific data exfiltration vectors
  • The exploit requires “always allow” permissions, allowing the deployment of arbitrary shell scripts and malware.

IBM’s Generative Artificial Intelligence (GenAI) tool Bob is susceptible to the same dangerous attack vector as most other similar tools: immediate indirect injection.

Indirect notice injection occurs when the AI ​​tool can read content found in other applications, such as email or calendar.



Leave a Comment

Your email address will not be published. Required fields are marked *