How did the government decide OpenAI’s frontier model was safe to release?

The U.S.

government cleared OpenAI’s frontier model Sol for public release, but no one outside a tight circle of executives and officials can explain how that decision was actually made.

A senior research analyst at Georgetown’s Center for Security and Emerging Technology says she has no visibility into those processes.

A former Trump policy advisor now working at OpenAI admitted publicly that nobody knows what the requirements are to get licensed.

Infighting and revolving-door personnel have left the evaluation process opaque, governed by personal relationships rather than transparent rules.

When OpenAI CEO Sam Altman listed the officials he spoke with — the Secretary of Commerce, the Treasury Secretary, the national cyber director — he provided no information about who tested the model or how they did it.