{
 "license": "CC-BY-4.0",
 "attribution": "Fide AI, Agentic Cyber Explorer",
 "url": "https://agentic-cyber-explorer.pages.dev/events/anthropic-cyber-jailbreak-severity-framework-2026/",
 "asOf": "2026-09-26",
 "id": "anthropic-cyber-jailbreak-severity-framework-2026",
 "date": "2026-07-02",
 "datePrecision": "day",
 "title": "Anthropic proposes Cyber Jailbreak Severity scale with Glasswing partners",
 "lane": "policy",
 "kind": "framework",
 "summary": "Anthropic published an early-draft Cyber Jailbreak Severity framework, developed with Project Glasswing partners, to score cyber jailbreaks on capability gain, breadth, ease of weaponization and discoverability, mapped to five levels from CJS-0 to CJS-4. It also described Fable 5's cyber classifier tiers, which block prohibited and high-risk dual-use requests such as exploit development while allowing defensive work like patching and incident response.",
 "whyItMatters": "A shared severity scale for safeguard bypasses is a precondition for proportionate government and industry responses like the June 2026 suspension.",
 "actors": [
  "anthropic"
 ],
 "topics": [
  "jailbreaks-and-safeguards",
  "access-controls",
  "standards-and-guidance",
  "incident-reporting"
 ],
 "atlas": [
  "model",
  "access-gate",
  "monitor"
 ],
 "artifacts": [
  "cyber-jailbreak-severity",
  "claude-fable-5"
 ],
 "sources": [
  {
   "url": "https://www.anthropic.com/news/fable-safeguards-jailbreak-framework",
   "publisher": "Anthropic",
   "title": "More details on Fable 5's cyber safeguards and our jailbreak framework",
   "date": "2026-07-02",
   "type": "primary",
   "accessed": "2026-09-25"
  }
 ],
 "keyFacts": [
  {
   "fact": "Severity bands: CJS-0 informational (0), CJS-1 low (1-3.5), CJS-2 medium (4-6.5), CJS-3 high (7-8.5), CJS-4 critical (9-10).",
   "locator": "CJS framework section"
  },
  {
   "fact": "Classifier tiers: prohibited use (blocked), high-risk dual use such as exploit development and privilege escalation (blocked), low-risk dual use (monitored, sometimes blocked), benign use (allowed).",
   "locator": "Classifier section"
  },
  {
   "fact": "Anthropic says Fable 5's safety margin was set larger than for other models, accepting more false positives.",
   "locator": "Classifier section"
  },
  {
   "fact": "Scoring axes: capability gain (0-4), breadth (0-2), ease of weaponization (0-2), discoverability (0-2).",
   "locator": "Cyber Jailbreak Severity section"
  }
 ],
 "significance": 3,
 "fideQuestions": [
  "FID-077"
 ],
 "methods": [
  "ai-assisted-exploitation",
  "ai-monitoring",
  "automated-patching",
  "injection-classifiers",
  "jailbreaking"
 ],
 "review": "assistant-drafted",
 "addedOn": "2026-09-25"
}