OpenAI banned accounts involved in activity that overlapped with publicly reported threat groups and displayed hallmarks consistent with PRC intelligence requirements, using AI to support phishing and scripting workflows.