Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
OpenAI is lagging behind The company on Friday confirmed the public launch of its next generation AI model, GPT-5.6, at the request of the Trump White House. OpenAI He said she would participate first Models With a small group of customers, which will be pre-approved by the US government. It will then work with management to slowly expand access.
OpenAI is not happy about this, according to a person familiar with the company’s thinking, but believes the delay and government approval process are only temporary. In a post on its blog, the company said it hopes to be able to make GPT-5.6 available to everyone in the coming weeks. OpenAI’s plans to delay the next generation of AI models at the request of the Trump administration were a first I mentioned By information.
“We do not believe that this type of government access process should become the default procedure in the long term,” OpenAI wrote in its blog post. “It preserves the best tools from users, developers, enterprises, cyber defenders, and global partners who need them. We are taking this short-term step because we believe it is the strongest path to broader availability in the coming weeks, as we work with the administration to develop the Cyber Executive Order framework and a repeatable process for future model releases.”
Earlier this month, President Trump signed Executive order Which aims to address cybersecurity concerns for powerful new AI models. The White House will establish a “voluntary process” for AI labs to share their models with the government 30 days before a broader release, the order said. The mandate included a carve-out, saying the US government would not turn its voluntary process into an actual licensing system for AI model releases. But in a press conference on Friday, OpenAI executives said no such voluntary framework exists yet. As a result, frontier AI labs are in a very strange interim period, where working with the US government on launching your own AI model does not seem voluntary.
The White House is asking OpenAI to delay the release of its AI models just two weeks after submitting them Direct export control to AnthropyWhich prompted the company to take its most advanced AI models offline for all customers. Anthropy’s dispute with the White House is Still unresolvedSome of the company’s employees are still prohibited from using more advanced AI models.
The Trump administration’s request for OpenAI and Anthropic to limit the availability of their most advanced AI models creates a problem An uncertain environment for other US AI labs. Over the past two years, the Trump administration has sought to remove regulations and bureaucracy that could stifle AI innovation in America and potentially hurt the country’s competitiveness with China. However, in recent months, the White House has grown increasingly concerned about the cybersecurity capabilities of new AI models, and has been quick to address the issue.
OpenAI plans to expand the pool of customers it can share GPT-5.6 with next week, including some international partners. OpenAI executives said they couldn’t share details about how the White House approved these clients, as the company only sends the U.S. government a list and then gets feedback on it, the executives said.
The White House did not immediately respond to WIRED’s request for comment.
OpenAI says its GPT-5.6 AI models will come in three flavors: Sol, the most capable version of the model; Terra, the middle-class version of the model; And Luna, a fast and affordable version. The company says the GPT-5.6 Sol is the most capable model to date on standards for testing cybersecurity, biology, and agent capabilities. Along with these new capabilities, OpenAI says it has a “multi-layered protection suite,” which aims to prevent bad actors from using its AI model for cyberattacks, among other malicious behavior.
