Trump Administration Mandates Vetting of OpenAI and Anthropic Customers for Top AI Models

The Trump administration has mandated that leading artificial intelligence companies, including OpenAI and Anthropic, submit their prospective customers for federal vetting before granting access to their most advanced AI models. The unprecedented intervention marks a sharp departure from the administration's previously hands-off approach to Silicon Valley regulation, signaling a new era of stringent oversight for frontier technologies.[1][3]

On Friday, OpenAI announced it would restrict the initial release of its new GPT-5.6 Sol model to a small group of "trusted partners" explicitly approved by the U.S. government. The company noted that the staggered rollout was executed at the administration's behest, with CEO Sam Altman expressing discomfort with the government "picking the customers."[1][2]

Hours later, the Commerce Department sent a letter to rival AI developer Anthropic, authorizing the redeployment of its powerful Mythos 5 model—but only to a restricted list of U.S.-based companies. Anthropic had previously been forced to pull Mythos 5 and Fable 5 offline entirely to comply with a sudden administration directive blocking foreign nationals from accessing the technology.[1][2]

The administration's sudden pivot stems from growing alarm over the cybersecurity capabilities of these frontier models. Officials, including Treasury Secretary Scott Bessent, have warned that systems like Mythos 5 and GPT-5.6 Sol are highly adept at discovering and exploiting vulnerabilities in global software infrastructure, prompting fears that they could be weaponized by malicious hackers or adversarial nations.[3][4]

The vetting process is being executed under the umbrella of an executive order signed by President Donald Trump earlier in June. While the order initially described a "voluntary" 30-day framework for the government to review advanced AI systems for national security risks, the strict customer-by-customer approval process imposed on OpenAI and Anthropic has surprised industry observers and effectively frozen the broader AI market.[4][5]

The ad-hoc nature of the export controls and vetting has drawn sharp criticism from both tech leaders and civil liberties groups. OpenAI stated that government access processes should not become the "long-term default," warning that the policy keeps vital tools away from legitimate cyber defenders and enterprise users.[2]

Meanwhile, digital rights organizations like the Electronic Frontier Foundation have condemned the administration's actions as opaque and unconstitutional. Critics argue that giving the government unilateral power to dictate customer access sets a dangerous precedent, leaving the policy vulnerable to political weaponization against uncooperative companies.[4]

The restrictions have also sparked anxiety among U.S. allies, who fear being cut off from cutting-edge American technology. While the Commerce Department's letter to Anthropic explicitly limited access to U.S.-based companies, OpenAI indicated it hopes to add foreign partners to its approved list in the coming weeks as it works with the White House to develop a more sustainable deployment framework.[1][2]

As the administration scrambles to formalize its AI governance strategy, the tech industry remains in a state of limbo. The White House has stated it is collaborating with frontier AI labs to develop shared approaches for scaling the technology, but until a clear regulatory framework emerges, the release of America's most powerful AI systems will remain tightly controlled by federal officials.[1][5]