White House asks OpenAI to limit its next model release

In a significant move, the White House has urged OpenAI to restrict the upcoming release of its GPT 5.6 model to a select group of government-approved partners. This request stems from concerns over the model's advanced capabilities, as reported by a source familiar with the discussions. The administration's request follows the imposition of an export control order on Anthropic, which resulted in the company retracting its latest models, Mythos and Fable. These models raised alarms in both Washington and financial markets due to their sophisticated cybersecurity features, which have led to fears of potential safety risks. According to the source, both OpenAI and the administration consider the new model to be comparable in capability to Anthropic's advanced offerings. In a bid to navigate the current regulatory landscape, OpenAI has agreed to limit the model's distribution, aiming for a public launch during a period characterized by a lack of federal regulatory standards for new AI technologies. A memo from OpenAI's CEO, Sam Altman, which was reported by The Information, indicated that the government is approving access on a case-by-case basis. "We’ve made it clear to the U.S. government that this is not our preferred long-term model, and we will work with them and other industry stakeholders to establish a more sustainable framework for future releases," Altman stated. A White House official confirmed ongoing collaboration with leading AI laboratories to devise shared strategies for addressing the challenges posed by scaling this type of technology. Although President Trump recently signed an executive order encouraging AI firms with advanced models to voluntarily submit them for government review 30 days prior to their release, the details of this framework remain unsettled. Currently, there is ambiguity among AI companies regarding which agency is responsible for AI regulation. While the White House made the request to OpenAI, the export control measures affecting Anthropic were issued by the Commerce Department. Experts assert that the government must be involved in discussions surrounding AI safety, particularly those that have implications for national security. However, the absence of a clear and consistent regulatory framework raises concerns that it could hinder industry growth. Brad Carson, head of Public First, a bipartisan pro-AI safety super PAC, pointed out the necessity for well-defined regulations. He remarked, "The Fable incident underscores the urgent need for clarity in regulations. At present, the approach appears ad hoc, personalized, and lacks transparency, which could lead to lawlessness. While it is indeed appropriate for the government to address dangerous products, including AI models, such actions must be executed transparently and fairly."