{
  "id": 16645,
  "url": "https://arxiv.org/abs/2601.23112",
  "title": "How Should AI Safety Benchmarks Benchmark Safety?",
  "summary": "arXiv:2601.23112v3 Announce Type: replace Abstract: AI safety benchmarks are pivotal for safety in advanced AI systems; however, they have significant technical, epistemic, and sociotechnical shortcomings. We present a review of 210 safety benchmarks that maps out common challenges in safety benchmarking, documenting failures and limitations by drawing from engineering sciences and long-established theories of risk and safety. We argue that adhering to established risk management principles, map",
  "authors": "Cheng Yu, Severin Engelmann, Ruoxuan Cao, Dalia Ali, Orestis Papakyriakopoulos",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-06T04:00:00.000Z",
  "fetched_at": "2026-08-06T05:10:11.148Z",
  "source_slug": "arxiv-cscy",
  "source_name": "arXiv cs.CY",
  "source_homepage": "https://arxiv.org/list/cs.CY/recent",
  "ethics_ai_record_url": "https://ethics.ai/record/16645",
  "original_url": "https://arxiv.org/abs/2601.23112",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}