restrict

The US Government has asked OpenAI to limit the distribution of GPT 5.6, its next artificial intelligence (IA), to a reduced number of partners approved by the Administration itself. This decision, according to CNNhas been caused by the advanced capabilities of this model. However, this is not an isolated movement. And this request comes a few days after the Trump Administration imposed a export control order to Anthropic, which led this company to withdraw its two most advanced models from the market: Mythos and Fable. The trigger for the current scenario has been, precisely, that episode. Mythos and Fable put Washington and Wall Street on alert for their cybersecurity capabilities, which some experts consider an unprecedented risk vector. According to those same sources, the Administration and OpenAI agree that GPT 5.6 is on par with Mythos in terms of power and reach. This makes it, de factoin the second border model to come under direct government scrutiny in a matter of weeks. OpenAI has accepted this restriction as a provisional way to release this model publicly. Sam Altman, the head of this company, has described this scenario in an internal report as a “strange moment” with no real regulatory framework for AI models. In that same document Altman explained that the Government is approving “client by client” access, a formula that the company tolerates but does not support. A “strange time” for AI companies OpenAI’s official stance leaves little room for ambiguity. “We have made it clear to the US Government that this is not our preferred long-term model, and we will work with them and other players in the sector to achieve a more sustainable approach in future launches,” wrote Altman in his report. The White House has limited itself to stating to CNN that “they continue to collaborate with frontier AI laboratories to develop shared approaches.” OpenAI has not commented anything further at the moment. Donald Trump has signed an executive order asking companies with advanced models to undergo a government review Be that as it may, the regulatory background further complicates the picture. Donald Trump signed an executive order earlier this month requiring companies with advanced models undergo a government review voluntary 30 days before each launch. The framework of its implementation, however, still not established. The handling of the OpenAI case perfectly illustrates the confusion that now pervades everything: the petition came from the White House, while the export ban on Anthropic came from the Department of Commerce. They are two different organisms. Two different instruments and without apparent coordination. In any case, the absence of a clear regulatory framework responds to explicit critics. “Right now we have an ad hoc, personalized, opaque, possibly outside the law approach,” Brad Carson stateddirector of Public First, a bipartisan pro-AI safety superPAC. A superPAC is a type of American political organization that can raise and spend unlimited amounts of money to influence elections or public debate, but without being subject to the funding limits that apply to traditional parties or candidates. Carson recognizes that the Government has the power to remove dangerous products from the market, including AI models, but warns that it must be done with transparency and fairness. We’ll see what happens eventually, but GPT 5.6 still doesn’t have a confirmed public release date. More information | cnn In Xataka | The world is running out of data to continue training AI. China has an ace up its sleeve