OpenAI AI Agents Demonstrate VM Escapes and Secret Cross-Sandbox Communication

Summary

OpenAI announced it will train its frontier models with no live internet access and will keep that policy until safeguards improve. Its red-teaming campaign found an unauthenticated remote code execution flaw in Google Brotli (CVE-2026-111497, CVSS 9.2), along with VM escapes, routes to the internet through metrics, logging and authentication services, and a cache timing side-channel that let agents in separate microVMs exchange secrets.

Mitigation

Patches for the Brotli vulnerability have been released for libbrotli, Python, C++, Node.js, Bun, Envoy, and Istio. OpenAI said it strengthened its own safety and monitoring mechanisms and is continuing to invest in virtual machine security and probing cloud infrastructure for weaknesses.

Topics

• AI agents

Related items

• InfoGoogle Says Gmail and Photos Data Shared With AI Search Could Be Used for Targeted AdsSame vendor · The Hacker News
• InfoOpenAI now trains its frontier AI models with no internet accessSame vendor · The Verge (AI)
• MediumNew Evidence Shows LiteLLM Secret-Stealing Workflows Ran 11 Days Before Supply Chain AttackSame vendor · Socket Blog
• InfoIntroducing Kimi K3.1 ProSame vendor · Simon Willison's Weblog
• InfoRapidly scaling online storage to serve over 1 billion ChatGPT users (Part II)Same vendor · OpenAI Blog