Wire · opportunities
Nvidia says its new platform can prevent AI agents from going rogue
◆ Sectors
◆ Geography
◆ Source
◆ Verified
Fusion42 · 29 September 2026 · Fusion42 review
Nvidia has launched the Open Agent Safety Platform to enhance AI security by preventing AI agents from escaping testing environments and accessing external systems. The platform includes Nvidia OpenShell to set boundaries on CPUs and Sentry on BlueField chips to quarantine unauthorized agent behaviour, collaborating with several major AI companies except OpenAI.
This Wire brief sits within Fusion42's coverage of AI & ML, and 13 sources have reported it between 28 Sep 2026 and 29 Sep 2026.
◆ ◆ The Wire takeaway
You must prepare for tighter AI security requirements as Nvidia’s platform sets new limits on AI agent actions to prevent breaches. This opens a door for AI startups to integrate stricter control measures or offer complementary security layers.
◆ Coverage
13 sources · first reported 28 Sep 2026 · latest 29 Sep 2026
◆ Related on Wire
◆ Topics