Nvidia Open-Sources OpenShell, a Kernel-Level Sandbox to Contain Rogue AI Agents
nordicinst · x · 2026-09-28
WIRED reports Nvidia is open-sourcing its answer to rogue AI agents after a string of incidents where agents hacked companies and probed US and Australian government sites.
- OpenShell, first shown at GTC in March, enters general release: a sandbox framework that isolates agent activity at the OS kernel level.
- Nvidia also launched the Open Agent Safety Platform; it says SpaceXAI uses it for Cursor agents and Grok models, and Anthropic is co-building security into Claude management.
- Dozens of partners are involved, including Anthropic, Cisco, CoreWeave, CrowdStrike, Dell, Hugging Face, JPMorganChase, Mistral, Microsoft, and Palantir.
Related event: NVIDIA Launches Open Agent Safety Platform with 100+ Partners(35 posts)→
More from Infra
- India's Sarvam launches Token Factory, hosting top open models on local infra — HarveenChadha · 2026-11-16
- Gemini 3.5 Flash now runs on air-gapped Google Distributed Cloud, offline up to 12 months — shashib · 2026-11-16
- Gemma.c: full Gemma 1/2/3 multimodal inference in a single C file, ~72 tok/s on CPU at INT8 — This-Purchase-3325 · 2026-11-16
- BIS Issues Interim Final Rule Addressing Remote Access to Advanced AI Chips — apples_jimmy · 2026-11-16
- Jailbroken PS4 runs a 9B open-source model at up to 43 tok/s by talking directly to its 2013-era GPU — AIFlow_ML · 2026-11-16
- Mobile AI Factory: agents auto-build on-device SLMs with 19x less GPU time — pmttyji · 2026-11-16