OpenAI banned accounts potentially associated with publicly reported DPRK-affiliated threat actors using AI to research intrusion tooling, phishing, malware, and cryptocurrency targeting.