paperAugust 2026Unreviewed

AI Security Leaderboard: Methodology, Results and Minimal Standard

Jasper Timm, Lukas Struppek, Ziwei Xu, Grace Cheong, Oscar Mata, Dan Zhao, Mick Yang, Isadora De Andrade, Xiaojun Jia, Yiming Li, Samuel Bauer, Heather McIntyre, Adam Gleave, Edward Yee, Kellin Pelrine

View resource PDF

Abstract

The AI Security Leaderboard is an independent benchmark that ranks the safeguards of frontier AI models from least to most secure. It tests models against the FAR$.$AI Minimal Standard for Safeguards, which represents a minimum bar for security: meeting it does not guarantee a secure model, but failing to meet it guarantees a lack of state-of-the-art security. Version 1.0 covers severe misuse requests across chemical, biological, radiological, nuclear, and explosive (CBRNE) threats and offensive

Cite

@misc{timm2026ai,
  title = {{AI Security Leaderboard: Methodology, Results and Minimal Standard}},
  author = {Jasper Timm and Lukas Struppek and Ziwei Xu and Grace Cheong and Oscar Mata and Dan Zhao and Mick Yang and Isadora De Andrade and Xiaojun Jia and Yiming Li and Samuel Bauer and Heather McIntyre and Adam Gleave and Edward Yee and Kellin Pelrine},
  year = {2026},
  month = aug,
  eprint = {2608.03070},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.03070}
}