Nvidia Announces New Platform to Restrict AI Agents
Nvidia has introduced the Open Agent Security Platform, which allows businesses to control autonomous AI agents and stop them if they exceed their authority.
Nvidia has announced the Open Agent Security Platform, combining software restrictions with an independent hardware watchdog to prevent artificial intelligence agents from exceeding their authority limits.
Security Layer for AI Agents
Nvidia has announced the new Open Agent Security Platform, which enables businesses to keep autonomous agents under control. This system makes it possible to stop agents even in cases where software guidelines fail.
Sentry Hardware Watchdog
The platform combines Nvidia OpenShell software with a separate watchdog called Sentry running on BlueField-4 data processing hardware. Sentry can intervene within milliseconds if agents go beyond their software boundaries.
Financial Sector and Partnerships
Financial institutions such as banks and payment providers can use this infrastructure to audit agents accessing sensitive data. Institutions such as Citi and JPMorganChase are working on open-source security technologies.
Support from Global Tech Companies
While it is stated that over a hundred organizations are working with the platform technologies, these partners include giant names such as Anthropic, Microsoft, Salesforce, SAP, IBM, and Palo Alto Networks.