Nvidia’s Answer to Rogue Agents Is an Open-Source AI Security System
In the wake of a series of high-profile AI safety incidents, Nvidia is introducing a new software tool that helps keep agents from escaping containment.
By Dillip Chowdary • Oct 01, 2026 • Source: Wired
Nvidia’s Answer to Rogue Agents: what actually changed
Wired reports: Nvidia’s Answer to Rogue Agents Is an Open-Source AI Security System. In the wake of a series of high-profile AI safety incidents, Nvidia is introducing a new software tool that helps keep agents from escaping containment.
Nvidia says SpaceXAI is using the Open Agent Safety Platform for its Cursor agents and Grok models. The company also says Anthropic and Nvidia are “building security into Claude Managed Agents.” Salesforce, Scale AI, and SAP are all confirmed to be integrating OpenShell to some degree.
Nvidia’s Answer to Rogue Agents: how it works
However, it is unclear whether OpenShell has been adopted by Nvidia’s full list of partners, or whether Nvidia is gesturing broadly.One notable name is missing entirely from Nvidia’s list: OpenAI. (Nvidia agreed to acquire Hugging Face earlier this month for $12.9 billion.)The chip giant has also developed a new software platform called Sentry, an isolated security domain for chips that’s supposed to continuously monitor long-running AI agents.
Advertisement
Tech Pulse Daily
Get tomorrow's pulse first
Join engineers who read Tech Pulse before stand-up. Free, weekday mornings.
Nvidia’s Answer to Rogue Agents: why it matters now
While Sentry is technically a software tool, it’s meant to be implemented on Bluefield, Nvidia’s line of programmable data processing units (DPUs). See the full write-up from Wired via the source link for quotes and complete context.
Nvidia’s Answer to Rogue Agents: who is affected
“With this, agents only have access to the intent that the security team wants them to have.”Boitano adds that Nvidia is working with both Arm and Intel to create a version of Sentry that works on the x86 chip architecture. The coalition’s stated goal was to reduce AI risks, particularly through a program called the Shared AI Findings Exchange (SAFE).
Nvidia’s Answer to Rogue Agents: what to watch
Last month, Boitano said that SAFE was designed to be “governed independently, with no single company or industry segment controlling its findings.”But one company keep
Developer Action Items
- ☐ Map where OpenAI / Anthropic / Claude sits in your stack (SDK, API key, billing, data-processing addendum).
- ☐ Hold the $12.9 billion figure to the primary report; do not brief a number that is not on the record.
- ☐ Hold non-urgent migrations until the integration or use-of-proceeds roadmap is public — day-one coverage is not a ship signal.
- ☐ If you are mid-contract or mid-POC, ask the vendor what changes for existing customers this quarter.
- ☐ Write the single decision this forces: stay, dual-source, or exit.
Author
Dillip Chowdary
Writes Tech Bytes coverage of AI, engineering, and the tools that actually ship. Editor of Tech Pulse Daily.
Related on Tech Bytes
When can we say AI made a scientific discovery?
Read →
Meta launches enterprise AI platform, hires MongoDB CEO to lead new initiative
Read →
Radicle Discloses Critical Flaws Exposing Private Repositories in Plain Text
Read →
Bose Launches New Wired Earbuds After More Than a Decade
Read →
Today's Tech Pulse briefing
Full briefing →
Advertisement