“OpenAI and its leadership have violated every major AI safety guardrail that was established by the research community a decade ago.”