Open Secure AI Alliance Proposes SAFE Guidelines for Agentic AI Cybersecurity

ny blog post detailing a new Request for Comments from the Linux Foundation on Shared AI Findings Exchange (SAFE), a set of proposed guidelines aimed at turning agentic AI cybersecurity incidents into shared protection across the industry, timed to coincide with the Black Hat conference in Las Vegas.

According to NVIDIA, the guidelines are being drafted by an Open Secure AI Alliance working group, which it said now numbers more than 120 organisations, including NVIDIA itself alongside Cisco, CrowdStrike, Hugging Face and Red Hat. NVIDIA said the proposals include confidentially collecting and analysing AI incidents and near-misses, notifying affected parties, identifying recurring control failures, and publishing evidence-based operating recommendations.

NVIDIA said the SAFE framework builds on a broader wave of open security tools contributed by alliance members across the AI security stack, including identity and access controls from Okta and Palo Alto Networks, agent harnesses from Amazon and Microsoft, and specialised security models from Cisco and CrowdStrike. NVIDIA said Mistral also released its Shieldstral safety classifier as open weights the same day. This is a small selection; NVIDIA's post lists roughly two dozen further member contributions across the stack.

NVIDIA said Amazon and Visa were among the newest alliance members, contributing tools including Strands Agents, Cedar, and the Visa Vulnerability Agentic Harness. Members are due to gather for a group photo at Black Hat on 4 August.

Who Owns AI Security in the Enterprise? Governance Is Still in Its Infancy

Who actually owns AI security in your organisation — and how mature is your governance around it? Two senior CISOs from vastly different environments give a straight answer: ownership sits with the CISO for now, and governance, even in well-run programmes, is still in its infancy. AI is shifting enterprise risk from defending infrastructure to defending decisions. Agentic AI operates semi- or fully autonomously, traditional security controls don’t fit probabilistic systems, and no single vendor covers the full attack surface. Speakers: Andy Holliday, CISO at Petrofac, Lester Godsey, CISO at Arizona State University and Stewart Tinson, Project Director, AI-360 You’ll learn: • Why the CISO is the only realistic owner of AI security risk for the next 5 years • Why agentic AI breaks deterministic security controls and what to do about it • How ASU built an actionable AI framework supporting 60+ large language models • Practical controls: API key hygiene, command whitelists, blast radius reduction • Why no single vendor can cover AI security end-to-end Key topics: Agentic AI risk • AI governance maturity • Threat model transformation • CISO ownership • Incident response for AI • Ethics & training data bias • Vendor landscape reality • Probabilistic vs deterministic controls For CISOs, CIOs, and risk leaders making decisions about AI adoption now.

BrightTALK