OpenAI Has Developed Its Most Advanced Model Yet. The U.S. Government Cleared Its Rollout

The U.S. government has cleared OpenAI to move ahead with a broad rollout of its GPT-5.6 family of artificial intelligence models, ending weeks of restricted access that had limited the technology to a small group of vetted partners during additional federal review.

The company is expected to begin the wider launch on Thursday after completing further testing and discussions with U.S. officials. The rollout follows a temporary delay requested by the federal government as agencies examined the capabilities of OpenAI’s most advanced models and their potential national security implications, CNBC reported.

The approval comes at a time when advanced AI systems are receiving heightened scrutiny because of their ability to assist with cybersecurity research, biological analysis, and sophisticated coding. Those capabilities have become increasingly important as governments weigh the benefits of rapid AI development against concerns that the technology could be exploited by hostile states or malicious actors during a period of rising geopolitical tensions.

OpenAI had originally introduced GPT-5.6 Sol, Terra and Luna in late June but limited access to what it described as a small group of trusted partners while complying with federal oversight. The company said at the time that it remained committed to broad public access and expected wider availability in the coming weeks.

The U.S. Department of Commerce has now cleared the broader release following additional testing and meetings between OpenAI and government officials, Axios reported, citing a source familiar with the matter.

The company confirmed late Tuesday that GPT-5.6 Sol, along with lower-cost models Terra and Luna, will launch publicly on Thursday. Reuters reported that GPT-5.6 Sol is OpenAI’s most advanced model to date, while Terra is designed for lower-cost enterprise workloads and Luna targets high-volume, faster processing tasks.

OpenAI had previously limited access to vetted partners whose identities were shared with U.S. authorities during the review process.

Federal oversight of advanced AI models has expanded significantly in recent months. Earlier this year, President Donald Trump signed an executive order establishing a voluntary framework that allows developers of certain frontier AI models to provide them to the U.S. government for review before broader deployment.

OpenAI’s rollout extended beyond that framework after federal officials requested additional evaluation before allowing wider access. Quartz reported that technical staff from OpenAI worked directly with the Commerce Department’s Center for AI Standards and Innovation during the review, responding to questions raised by government officials before approval was granted.

The government’s closer involvement reflects growing concern over the strategic importance of advanced AI technologies as the United States and China compete to develop increasingly capable systems.

That scrutiny has affected other leading AI developers as well.

Last month, Anthropic temporarily withdrew public access to its Claude Fable 5 and Mythos 5 models after a Commerce Department export control order linked to national security concerns. Restrictions on Fable 5 were lifted last week after safeguards were implemented, although Mythos remains available only to selected U.S. organizations, Reuters reported.