SL5 Standard for AI Security
A NIST SP 800-53 overlay for frontier AI infrastructure achieving nation-state-level security. 43 controls across 10 families by the Security Level 5 Task Force.
Version 0.1 - Long lead times An integrated security control profile for frontier AI infrastructure, anchored in NIST SP 800-53 and extended with supply-chain and facility requirements, targeting nation-state-level security by 2028/2029. This version 0.1 focuses on long lead time controls, prioritizing measures that must be planned years in advance. Future versions will extend to shorter-lead-time measures and fine-grained controls. 60 controls across 12 families - March 2026 (updated June 23, 2026) About This Document Security Level 5 (SL5) is a security posture for AI systems that…
saved by
related reading
- Strategic AI Sabotage: State Attacks on Advanced Systems' Developmenttwmstone.com
- Strategic Visionsstatic1.squarespace.com
- Project MoAT - Mock SL5 Datacenter Buildsl5.org
- Off Target | CNAScnas.org
- Who’s doing what on AI security in the US government?horizonlaunchpad.substack.com
- The End-State Fallacy: Where Is AI Security Headed?endstatefallacy.com
- America’s AI Action Planwhitehouse.gov
- Frontier safety blueprintcdn.openai.com
- Promoting Advanced Artificial Intelligence Innovation and Security – The White Housewhitehouse.gov
- Common Elements of Frontier AI Safety Policiesmetr.org
- Anthropic's Responsible Scaling Policy \ Anthropicanthropic.com
- AI Integrity: Defending Against Backdoors and Secret Loyalties - Institute for AI Policy and Strategyiaps.ai