OpenAI wins clearance to broadly launch GPT-5.6 as White House disputes the 'green light'

OpenAI will broadly launch its GPT-5.6 model family — Sol, Terra and Luna — starting Thursday, July 9, after US regulators completed additional security testing and cleared the models for wider release. The rollout reverses a distribution limit imposed less than two weeks earlier.

The clearance reportedly followed extra evaluations by the Commerce Department’s Center for AI Standards and Innovation. OpenAI has not published its own post detailing the decision, and how binding the government’s sign-off ever was is now contested.

On June 26, OpenAI restricted GPT-5.6 to a small group of trusted-partner organizations at the US government’s request while security reviews were underway, according to OpenAI’s deployment safety page. Thursday’s rollout removes that cap for the general public.

The family spans three tiers. Sol, the flagship, targets complex coding and security research at $5 per million input tokens and $30 per million output tokens. Terra is a mid-tier model priced at roughly half the cost of GPT-5.5. Luna is the fastest and lowest-cost option.

The “clearance” framing is disputed. A White House official told reporters that deployment timing “rest[s] entirely with the compan[ies]” and that “no such permission is required or granted.” That account sits awkwardly against the reported role of the Commerce Department review, leaving it unclear how much Washington shaped the timeline.

Either way, the episode is a rare case of a frontier model held back from public release on security grounds and then let out. Regulators and rivals will study the sequence as a template for how far the US government can slow a launch.