OpenAI limits GPT-5.6 rollout after US government request

OpenAI is limiting the release of its newest AI models — the GPT-5.6 lineup — to a “small group of trusted partners” at the request of the U.S. government, the company said Friday. The lineup includes Sol, OpenAI’s flagship model; Terra, a more balanced model for everyday use; and Luna, a faster, lower-cost option. Although Sol is the most powerful of the three, the Trump administration restricted the release of all three, and OpenAI said the preview is limited to partners “whose participation has been shared with the government.”

The request follows a broader pattern of the U.S. government pressuring AI companies to restrict their most advanced systems. After Anthropic released its most powerful public model, Fable 5, the administration ordered the company to remove access for any foreign national — prompting Anthropic to take the model down entirely. Dean Ball, a former White House AI adviser who is soon to join OpenAI, says President Trump’s recent executive order — which asks certain AI companies to voluntarily submit their most advanced models for government review up to 30 days before release — has created a de facto involuntary licensing regime for frontier AI. Without clearly defined government safety standards, Ball argues, the result could be endless launch delays that might give China a hand in the AI race and jeopardize the billions of dollars going into AI infrastructure buildouts.

OpenAI complied but made clear it wasn’t happy with the arrangement. “We don’t believe this kind of government access process should become the long-term default,” the company said in a Friday blog post. “It keeps the best tools from users, developers, enterprises, cyber defenders, and global partners who need them.” OpenAI called the preview a “short-term step” that will put GPT-5.6 on the path to broader availability in the coming weeks, as it works with the administration on a new executive order framework for cybersecurity and a “repeatable process for future model releases.”

On specs, OpenAI says Sol is its strongest model yet, with improved agentic capabilities in coding, biology, and cybersecurity, plus a new “max” reasoning-effort mode and an “ultra” mode that uses coordinated subagents on highly complex tasks. Sol is slightly better at coding workflows than Anthropic’s Claude Mythos 5 — also effectively banned by the administration this month — and is competitive with the Mythos preview while using a third of the output tokens. Its safety guardrails are built directly into the model’s core behavior rather than a separate filter on top, echoing the downrouting problems that drew backlash when Fable 5’s classifiers rerouted high-risk prompts to an older model. GPT-5.6 stays limited to select partners for now, but OpenAI plans to bring it more broadly to ChatGPT, Codex, and API users soon. Pricing: Sol costs $5 per million input tokens and $30 per million output tokens, Terra costs half that, and Luna costs $1 and $6, respectively.

Why it matters

The episode raises questions about how much authority governments should hold over frontier AI releases. Without clearly defined safety standards, Ball warns, repeated delays could hand rivals like China an edge in the AI race and put the billions of dollars flowing into AI infrastructure at risk.