Become a Member

Regulation TrackerLaws, orders and treaties read for their security duties and clocks Standards & FrameworksAttack taxonomies, control catalogues and frontier frameworks AI Incident DatabaseThe security incident record, with mechanism and lesson Research Paper LibraryThe papers that define the field, summarised OSS DirectoryRed teaming, guardrail, supply chain and evaluation tools Case StudiesProgrammes built, with measured outcomes AI Vulnerability TrackerCVEs and advisories for the components AI systems are built from

Latest in the tracker

New York Stop Addictive Feeds Exploitation (SAFE) for Kids Act and implementing rulesNew York's law restricting algorithmic feeds for minors takes effect on 25 January 2027 under final rules published in July 2026, with certified age assurance and $5,000-per-violation penalties.

Knowledge Base Regulation Tracker Standards & Frameworks OSS Directory AI Incident Database Research Paper Library Case Studies Society Articles News Jobs About Charter Become a Member

Home/Standards & Frameworks/Securing AI Model Weights: Preventing Theft and Misuse of Frontier Models

Securing AI Model Weights: Preventing Theft and Misuse of Frontier Models

RAND Corporation report defining five security levels (SL1-SL5) for protecting frontier model weights against a spectrum of attackers

The reference framework for model-weight security: 38 attack vectors, five attacker capability categories and five security levels with recommended controls, adopted by frontier developers' safety frameworks (ASL security standards) and by the NIST misuse-risk guidance.

model weightssecurity levelsrand

Sources

rand.orghttps://www.rand.org/pubs/research_reports/RRA2849-1.html

Contribute

Spot an error, or know a record we should add? Members can propose additions and submit corrections.

Add standard Report a correction or request removal