OpenAI has announced that its new AI model, ChatGPT 5.6, will not be available to everyone at once. According to a report from The Information, the company plans a staggered release, with the first users being only those approved by the federal government. This marks a significant shift from previous launches, which often opened to the public in beta phases.
Controlled access for national security reasons
In an internal memo to employees, CEO Sam Altman explained that federal leaders will approve access customer by customer during the preview period. Only after a couple of weeks, he added, will a more general release of the 5.6 model follow. 'We have made clear to the US government that this is not our preferred long-term model and will work with them and others in industry to achieve a more sustainable approach for future releases,' Altman reportedly said in the memo.
Sponsored Protocol
Involvement of multiple federal agencies
Several agencies appear to have driven this course change. The Information cited interactions with the Office of the National Cyber Director and the White House Office of Science and Technology Policy, as well as involvement from Commerce Secretary Howard Lutnick. Neither the White House nor the Office of the National Cyber Director responded to requests for comment. This development follows an executive order signed by President Donald Trump earlier this month, asking AI companies to undergo a voluntary federal review of their most powerful models before public release. The government aims to create a framework to standardize the assessment of new models.
Comparison with Anthropic's move
Recently, OpenAI rival Anthropic disabled access to two of its models after a federal directive. The order provided no specifics about security concerns but blocked access for foreign nationals. This episode, along with OpenAI's additional release phase, highlights the ongoing confusion about how the review process will work and how voluntary it truly is. This move by OpenAI could affect user trust in AI code generation, as already seen in the case of SpaceX's acquisition of Cursor, which raised questions about model transparency and reliability.
Sponsored Protocol
For more insights on how trust in generative AI is evolving, read our article on SpaceX and Cursor. Additionally, for the global regulatory context, refer to the Wikipedia page on AI safety.
