LLM agents developed persistent competitive strategies in repeated prisoner's-dilemma games
Review RAT-166 · Catalogue RAT-166
Original paper evaluates iterated-prisoner-dilemma tournaments among models and coded strategies. Cooperation/defection is game behavior, with no organizational security victim.
Reviewed: · Rule: roguewatch-organizational-harm-v1
Dates and discovery
- Incident date
- Not disclosed
- Added to research / record
- Posted to RogueWatch
- No linked public incident
- First observed
- 2026-10-30T13:13:41.792482Z UTC
- Discovery channel
- User-supplied comparison audit
- Discovery source
- https://rogueaitracker.com/incidents/llm-agents-developed-persistent-competitive-strategies-in-re/
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
- Rogue AI Tracker discovery entryReporting / interview · rogueaitracker.com
- arxiv.orgOriginal account · arxiv.org
Decision history
- · Case-by-case comparison audit
RAT-166: out-of-scope. Original paper evaluates iterated-prisoner-dilemma tournaments among models and coded strategies. Cooperation/defection is game behavior, with no organizational security victim. Source checks: read: https://arxiv.org/abs/2507.02618. Original paper evaluates iterated-prisoner-dilemma tournaments among models and coded strategies. Cooperation/defection is game behavior, with no organizational security victim.
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.