Claude Code: About 30 targeted organizations in espionage campaign
Review CAT-X07 · Catalogue X07, RAT-037
EXCLUDED — humans chose targets and goals. Human operators selected targets and intervened at decision points. Important near-match, but not an AI independently initiating unauthorized targeting.
Reviewed: Not recorded · Rule: external-intrusion-v1
AI did 80-90% of the work but humans directed it; out of scope for the rogue-model seed. Candidate for an agentic layer.
Reviewed: Not recorded · Rule: roguewatch-organizational-harm-v1
Evidence status
EXCLUDED — humans chose targets and goals
Dates and discovery
- Incident date
- Detected mid-September 2025; exact start undisclosed
- Catalogue disclosure / source date
- Added to research / record
- Posted to RogueWatch
- No linked public incident
- First observed
- Not recorded
- Discovery channel
- Not recorded
Dates use America/Los_Angeles. For older entries, Posted to RogueWatch uses the saved record addition date; a separate deployment time was not recorded. Future posting timestamps record first public qualification. Source publication and incident occurrence are separate.
Sources
- Anthropic — Disrupting the first reported AI-orchestrated cyber espionage campaign (Nov 13, 2025)Original account · www.anthropic.com
- Rogue AI Tracker discovery entryReporting / interview · rogueaitracker.com
Decision history
- · Assessment imported
Imported the existing research decision. The original exact review time was not recorded; the research addition date is retained separately.
Evidence source - · Case-by-case comparison audit
RAT-037: out-of-scope. Anthropic primary explicitly says state-sponsored attackers selected targets and built the attack framework. Claude executed their intended espionage campaign. Same excluded catalogue X07. Source checks: read: https://www.anthropic.com/news/disrupting-AI-espionage. Anthropic primary explicitly says state-sponsored attackers selected targets and built the attack framework. Claude executed their intended espionage campaign. Same excluded catalogue X07.
Evidence source - Not recorded · Retained review-queue assessment
AI did 80-90% of the work but humans directed it; out of scope for the rogue-model seed. Candidate for an agentic layer.
Evidence source
A review decision applies only to the named scope. Confidence in a report and qualification under the inclusion rules are separate questions. Review records do not add incidents to the globe or leaderboard.