AI agents behind first-of-its-kind incident traced to Tencent-run evaluations
dylfreed · x · 2026-10-14
Reporter Dylan Freedman says the AI agents behind a first-of-its-kind incident came from evaluations run by Chinese tech giant Tencent, based on an analysis of network activity and a person close to Tencent.
The researchers' full report on the incident is now available. It marks the first public attribution of such an agent incident to a specific company, raising questions about how AI agent evaluations are sandboxed and audited.
Related event: First out-of-control AI agent incident traced to Tencent internal testing(4 posts)→
More from Safety
- AI auditing scholar Deb Raji joins Stanford HAI to tackle who's liable when AI causes harm — StanfordHAI · 2026-10-15
- Anthropic pays $3,850/week plus $15K monthly compute for 4-month AI safety Fellows program — AIFlow_ML · 2026-10-15
- Denmark moves to give people legal control over deepfake likenesses — RosalindPicard · 2026-10-15
- moyix on AI agents hunting bugs: they stop exploring after one working exploit path — moyix · 2026-10-15
- Rivian owner spots ~8GB of weekly uploads, questioning car data collection — generativist · 2026-10-15
- Welsh politician caught using smart glasses as teleprompter in Senedd debate — nordicinst · 2026-10-15