OpenAI claims it used its own model to mediate an insider threat concern, but some are questioning the details of this incident.