Topic
Gen Threat Labs
Latest news
- 23 OctResearch shows AI agents can be manipulated by untrusted content
Gen Threat Labs demonstrated in a controlled experiment that AI agents can treat untrusted content as instructions, execute system commands, transmit data and perform destructive file operations. The report and related findings also highlight malware targeting assistant configurations and credentials, raising practical security concerns as agents gain more autonomy.
Security · 1 source
Articles
No articles about Gen Threat Labs yet.