American chipmaker Nvidia has introduced a security platform for AI agents — programs capable of carrying out tasks on their own. It lets users specify in advance which data and services an agent is allowed to access. If the agent tries to cross those boundaries, the system can, according to the company, isolate it within milliseconds, the website infohub.kz reports.
More than 100 organizations are working with the platform's technologies.
"AI's enormous potential will only benefit society if we solve the issue of its safety," said Nvidia founder and CEO Jensen Huang.
The Nvidia Open Agent Safety Platform brings together two tools. OpenShell sets the rules: which data, programs and external systems an agent may access while it works. This open-source software is already available to developers. It can be used on Nvidia Vera processors and adapted for hardware from other manufacturers.
The second tool, Sentry, runs on a separate Nvidia chip and monitors the agent's actions independently of the main program. If the agent tries to bypass the restrictions, Sentry is supposed to send it into "quarantine" and stop it. Nvidia describes this capability as part of a system project based on BlueField-4 chips.
Among the organizations using or developing the platform's technologies, Nvidia named Anthropic, SpaceXAI, Salesforce, JPMorganChase and Citi. Anthropic is working with the company on additional access restrictions for enterprise Claude agents. SpaceXAI is applying the platform to Cursor agents, which help write code, and to Grok models.
The impetus was a series of cases in which AI agents began slipping out of control. In July, OpenAI reported that its models, during a cybersecurity skills test, bypassed restrictions meant to block their internet access. The agents penetrated some OpenAI systems and servers of the Hugging Face platform, where developers host AI models. This happened during testing but affected real systems.
Anthropic reported three more cases: Claude models, during test assignments, gained unauthorized access to the systems of three organizations. According to the company, the agents believed they were working in a closed training environment, even though internet access was in fact open.
Earlier, Kursiv wrote that an OpenAI AI agent gained unauthorized access to Australia's Medicare health portal. The incident occurred in June, but the company notified the government about it three months later.