GoKawiilNvidia releases OpenShell AI-agent security sandbox for general use
Nvidia has moved its OpenShell agentic AI sandbox, first unveiled at GTC in March, into general availability for all users. The tool isolates AI agents' activity at the operating system kernel level while they perform tasks, and Nvidia says it has security collaborations with companies including Anthropic, Cisco, CoreWeave, Dell, Microsoft, Palantir, Salesforce and SAP.
GoKawiil's interpretation of the reporting above, not reported fact.
The rollout comes amid disclosed incidents of AI agents hacking companies and probing government websites, suggesting growing industry pressure to contain autonomous agent behavior before deployment at scale. Nvidia's push to rally dozens of partners around an open-source standard could position it as a central gatekeeper for AI agent safety infrastructure, though the notable absence of OpenAI from the announcement—despite both companies confirming involvement—raises unanswered questions about the partnership's actual scope.
- OpenShell, Nvidia's kernel-level AI agent sandbox, is now in general release for all users.
- Nvidia lists security collaborations with Anthropic, Cisco, CoreWeave, Dell, Microsoft, Palantir, Salesforce, SAP and others.
- OpenAI was omitted from the announcement despite both firms confirming its involvement, and neither explained why.
Source: wired.com — Lauren Goode, 2026-09-28
Published there as: “Nvidia’s Answer to Rogue Agents Is an Open-Source AI Security System”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.