The Linux Foundation released a Request for Comments document on August 4, 2026, outlining a proposed framework for agentic AI cybersecurity transparency, as the annual Black Hat conference opened in Las Vegas that same day.
The framework is called Shared AI Findings Exchange, or SAFE. According to NVIDIA, the guidelines are designed to strengthen cybersecurity transparency for agentic AI systems, and members of the Open Secure AI Alliance are actively developing them.
The Open Secure AI Alliance counts more than 120 member organizations, NVIDIA said.