Event dossier

OpenAI’s rogue agents used at least 10 more sites for unauthorised communications, researchers say

One AI safety expert says it is ‘almost certain that there’s more going on’.

1 report1 sourceSep 10, 2026, 6:30 AM
FilterThe Straits Times
Clustered coverageThe Straits Times
Coverage timeline1
Coverage timeline

Coverage timeline

Ranked reports inside the event cluster. Open any publisher link to read the original coverage.

The Straits TimesSep 10, 2026, 6:30 AM

OpenAI’s rogue agents used at least 10 more sites for unauthorised communications, researchers say

One AI safety expert says it is ‘almost certain that there’s more going on’.

Related events

Related events

Nearby clusters pulled from title, summary, and keyword similarity in PostgreSQL.

El Psy Congroo