{
 "license": "CC-BY-4.0",
 "attribution": "Fide AI, Agentic Cyber Explorer",
 "url": "https://agentic-cyber-explorer.pages.dev/events/deepmind-frontier-safety-framework-v3-1-2026/",
 "asOf": "2026-09-26",
 "id": "deepmind-frontier-safety-framework-v3-1-2026",
 "date": "2026-04-17",
 "datePrecision": "day",
 "title": "DeepMind Frontier Safety Framework v3.1 raises security for its cyber critical capability level",
 "lane": "policy",
 "kind": "framework",
 "summary": "Google DeepMind's Frontier Safety Framework version 3.1 introduced Tracked Capability Levels for earlier warning in CBRN and ML R&D and misalignment, and raised the recommended security for the CBRN, cyber and harmful manipulation CCLs to Security Level 2+. The cyber CCL definition itself, Cyber uplift level 1, is unchanged from v3.0.",
 "whyItMatters": "It raises the recommended weight security for cyber-capable models to Security Level 2+, adding protection against non-state and insider theft, while keeping the 2025 rationale that higher security levels are likely not warranted.",
 "actors": [
  "google-deepmind"
 ],
 "topics": [
  "capability-thresholds"
 ],
 "atlas": [
  "model"
 ],
 "artifacts": [],
 "sources": [
  {
   "url": "https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/strengthening-our-frontier-safety-framework/frontier-safety-framework_3-1.pdf",
   "publisher": "Google DeepMind",
   "title": "Frontier Safety Framework Version 3.1",
   "date": "2026-04-17",
   "type": "primary",
   "accessed": "2026-09-25"
  }
 ],
 "keyFacts": [
  {
   "fact": "Version 3.1 (April 17, 2026) raised security for CBRN, Cyber and Harmful Manipulation CCLs to Security Level 2+ to protect against non-state actors and insider threats.",
   "locator": "Section 5.3 Past Updates and Changes"
  },
  {
   "fact": "In v3.1, Table 2.2.2.a pairs Cyber uplift level 1 with Security level 2+ (v3.0 said Security level 2).",
   "locator": "Section 2.2.2, Table 2.2.2.a"
  },
  {
   "fact": "TCLs were introduced for CBRN and for ML R&D and misalignment, not for cyber.",
   "locator": "Section 1.2"
  }
 ],
 "significance": 3,
 "fideQuestions": [],
 "methods": [],
 "review": "assistant-drafted",
 "addedOn": "2026-09-25"
}