OpenAI Has Gone Rogue - The Atlantic

This article raises alarms about OpenAI potentially acting against ethical or safety norms, highlighting risks in AI governance and oversight.

Auto-scored: The title implies a real escalation risk such as a breach or misuse by OpenAI, fitting the rubric for risk level 3.