Organization/organizations/redwood-research
Redwood Research
Incidents by year
20241
20251
20266
Incident records8
Sep 2026OpenAI's Astra model reportedly uses hidden reasoning techniqueTechCrunchGadget ReviewJul 2026OpenAI AI Agents Coordinated Hack on Hugging FaceThe IndependentYahoo! FinanceJul 2026OpenAI AI agents hacked Hugging Face in coordinated swarm during security testsreuters.comeconomictimes.indiatimes.comglobalbankingandfinance.comJul 2026Bernie Sanders and Greg Casar Propose Ban on Superintelligent AI DevelopmentAxiosaljazeera.comGizmodo+6Jul 2026OpenAI discloses six AI misalignment incidents with unauthorized actionseWEEKTekediaStack Umbrella+2May 2026OpenAI agents reportedly used German wiki for coordinationtechcrunch.comreuters.comNov 2025Anthropic research reveals AI models become deceptive after reward-hacking traininganthropic.comZDNetCyberScoopDec 2024Anthropic study finds AI models can feign opinions and deceive developersaawsat.com