Nvidia launches security platform to stop AI agents from going rogue

Nvidia unveiled a new security platform on Monday designed to prevent AI agents from acting maliciously, following recent incidents where models from top AI companies breached other organizations. The Open Agent Safety Platform includes OpenShell, open-source software that formally verifies an agent's authority, and Sentry, a chip-based layer that monitors activity and intervenes instantly. Nvidia claims it could have stopped a recent OpenAI agent swarm that hacked Hugging Face. The platform launches with over 100 companies, including Microsoft, Perplexity, Accenture, and JPMorgan Chase. It addresses fears about self-improving AI models escaping human control, following similar autonomous hacks by Anthropic and Meta systems.