OpenAI's GPT-5.6 Limited After Trump Admin Request
Get the Finance newsletter
Daily finance — markets, central banks, M&A, the prints that move money. Free.
- OpenAI launched a limited preview of GPT-5.6 Sol (flagship), Terra (balanced), and Luna (fast/affordable), its next-generation model lineup announced on June 26, 2026
- GPT-5.6 Sol matches Mythos Preview on ExploitBench, adds Ultra mode with subagents for complex workflows, and includes a max reasoning setting for deep problem-solving, per OpenAI's announcement
- The Trump administration requested that OpenAI limit the rollout, resulting in restricted access for select partners and government-vetted users, according to CNN and PYMNTS
- OpenAI pushed back on the limits, saying restrictions "shouldn't be the norm" and that the government-controlled access model is unsustainable, per TechCrunch and The Decoder
- Neowin reported GPT-5.6 Sol beats Claude Mythos 5, while The Verge framed the launch amid "US AI regulatory drama" — coverage split between capability claims and the government's role
- Forbes and Constellation Research noted the release is pending US government sign-off, underscoring that even the preview tier requires federal approval before broader access
Why it matters: The White House successfully pressured OpenAI to gate its most advanced model behind a vetting process — a precedent that puts frontier AI deployment under de facto executive branch control without new legislation, while the model's own capabilities (matching Mythos Preview on ExploitBench, new subagent Ultra mode) get sidelined in the political story.
