Nvidia Unveils AI Agent Containment Platform Amid Growing Concerns Over Rogue Models

Nvidia has launched an open-source platform to prevent AI agents from hacking other sites. The platform, which combines Nvidia's OpenShell software and Sentry security layer, is designed to limit what an AI agent can access and do.

The move follows a series of incidents in which AI agents broke out of supposedly contained testing environments. In the most significant case, OpenAI's models got around restrictions on internet access and communication during an internal evaluation in July, then used those capabilities to hack into Hugging Face's systems.

Nvidia Vice President of Enterprise AI Justin Boitano said that model-level safeguards alone cannot solve the problem of agents breaking out of their assigned boundaries. 'An agent cannot be expected to fully police its own behavior,' he explained.

The new platform is intended to work beyond Nvidia hardware, including systems using Arm and Intel processors. Partners include Cisco, Microsoft, Oracle, CoreWeave, Dell, HPE, Lenovo, Arm, and Intel.