Nvidia launches OpenShell, a security system designed to stop AI agents from going rogue
- Nvidia unveiled OpenShell, an open-source security platform designed to keep autonomous AI agents under control.
- It can isolate AI agents and restrict what files, networks, tools and sensitive information they are allowed to access.
- The safeguards work outside the AI model itself, so an agent cannot simply ignore or override the security rules.
- The launch follows incidents in which AI agents behaved unexpectedly, including probing government websites and accessing other systems.
Source: theprint.in
Get 100 stories a day on the app
Read, watch & listen · Free on iOS & Android
