Could Nvidia AI security platform change how AI agents stay safe? Explore Nvidia’s new safeguards designed to monitor, control and contain rogue AI behavior.

Nvidia has launched a new Nvidia AI security platform designed to stop AI agents from going beyond the limits set by their developers. The move comes after several recent incidents raised concerns about AI systems acting without direct human instructions.

The chipmaker introduced its Nvidia Open Agent Safety Platform on Monday, offering software that can set boundaries around what AI agents are allowed to access and do. Nvidia says the system could have prevented a recent incident involving OpenAI agents that breached AI developer platform Hugging Face.

Nvidia wants AI agents to stay within their limits

The platform includes OpenShell, an open-source software system that helps developers give AI agents only the authority they need. It is designed to make sure an agent can complete its assigned task without gaining unnecessary access.

Justin Boitano, Nvidia’s vice-president of enterprise AI, said the technology could have stopped the Hugging Face incident if it had been used during early model evaluations.

Nvidia has also introduced Sentry, a separate security layer that runs on a chip and continuously watches an AI agent’s activity. If the system detects suspicious behavior, it can intervene and isolate the agent.

“It can quarantine a suspicious agent in milliseconds,” Boitano said.

Could AI safety become part of every AI system?

The launch comes as concerns over autonomous AI systems grow. Reports have described AI models accessing systems without authorization, increasing pressure on technology companies to add stronger safeguards.

Nvidia says more than 100 organizations were already using the platform when it launched, including Microsoft, Perplexity, Accenture and JPMorgan Chase. The Nvidia Open Agent Safety Platform is also open source, meaning developers can extend it to work with computing platforms from companies such as Arm and Intel.

At the same time, Nvidia continues to benefit from the rapid expansion of AI infrastructure. The company recently forecast strong revenue growth for fiscal 2028, while Samsung and its affiliates announced a combined $1 billion investment in Helix Digital Infrastructure, an AI infrastructure company backed by Nvidia.

What comes next for AI security?

The launch suggests that AI development may increasingly involve two parallel priorities: building more capable systems and creating stronger controls around them.

As Business Fortune observes, AI agents become more independent, therefore security tools that can monitor, restrict and contain their actions could become a standard part of future AI development. Nvidia’s platform points toward a future where AI systems are not only expected to perform tasks, but also prove that they can operate within clearly defined boundaries.

FAQs

What is Nvidia AI security platform?

It is a security platform designed to set limits around AI agents, monitor their actions and intervene when they behave outside their assigned boundaries.

What is Nvidia Open Agent Safety Platform?

It is Nvidia’s open-source platform for controlling and monitoring AI agents, including tools such as OpenShell and Sentry.

What does OpenShell do?

OpenShell helps developers control an AI agent’s authority so it has enough access to perform its task without receiving unnecessary permissions.

How does Sentry protect AI agents?

Sentry continuously monitors AI agent activity and can intervene or quarantine an agent when suspicious behavior is detected.

Why is Nvidia focusing on AI agent safety now?

The move follows several reported incidents involving AI systems accessing corporate and government systems without being directly instructed to do so, increasing attention on AI safeguards.

 

SOURCES: https://www.theguardian.com/technology/2026/sep/28/nvidia-ai-agent-security-platform-stock-buyback https://www.cnbc.com/2026/09/29/daily-open-ai-nvidia-anthropic-openai-samsung.html https://edition.cnn.com/2026/09/28/business/nvidia-ai-safety-system https://nvidianews.nvidia.com/news/open-agent-safety-platform