The Trump administration has taken new steps to assert more control over the rollout of future artificial intelligence model releases by dictating which companies and entities are allowed access to the latest frontier models, two people familiar with the matter told CNBC.
Until now, that decision was in the hands of American AI giants.
Both Anthropic and OpenAI have decided which companies and agencies have access to their most powerful models, and have often included major enterprise customers.
Anthropic unveiled its most capable Mythos cybersecurity model to a handful of partners with Project Glasswing. OpenAI was asked by the administration to gate its recent GPT-5.6 release, and has a similar consortium called Daybreak for its cybersecurity model.
A White House official told CNBC that it doesn't provide approvals for AI releases from private companies.
The official said any engagements, testing or meetings with government experts are "voluntary" and that "decisions on timing and scope of releases rest entirely with the companies," referring CNBC to Trump's recent executive order.
"The Administration continues to collaborate with all of America's frontier labs to strengthen the security of this technology without stifling innovation," they wrote.
However, last month the Trump administration blocked Claude Mythos 5 and Fable 5 due to "national security concerns," reinstating access after weeks of intense negotiations with Anthropic. OpenAI last month said it would limit new AI models to "trusted partners" to comply with government requests.
The White House is walking a fine line on regulation at a time when sophisticated AI tools pose massive cybersecurity risks and cheaper, open-weight models from China are quickly closing the gap with American frontier labs.
Chinese startup Moonshot AI unveiled its Kimi K3 model on Friday, which largely caught up to the performance of Fable and GPT-5.6, and even outperformed the U.S. frontier models in at least one independent benchmark.
David Sacks, founder of Craft Ventures and the former White House AI czar, called the Kimi breakthrough "concerning."
"This is how you lose the AI race," he wrote. "The rest of the world won't play by our rules if we bog ourselves down."
The administration has already taken several steps to reshape AI oversight in recent months, starting with President Donald Trump's June executive order, which asked companies to voluntarily give the government early access to models for testing.
This week, the administration launched its own program, dubbed "Gold Eagle," aimed at collaborating with the private sector to find and fix cyber vulnerabilities.
The so-called clearinghouse would put the White House in charge of greenlighting which companies can access new AI models, according to a person familiar with the matter, who spoke on condition of anonymity in order to discuss information that is not public.
The administration's moves have left the future of company-led initiatives such as Project Glasswing and OpenAI's Daybreak in doubt.
Going forward, according to one person, these rollouts will require explicit government approval for which partners can be involved.
— CNBC's Megan Cassella and Ashley Capoot contributed reporting.
Facts Only
* The Trump administration is dictating access to future artificial intelligence model releases.
* Anthropic and OpenAI previously controlled access to their most powerful models.
* Anthropic released the Mythos cybersecurity model to partners via Project Glasswing.
* OpenAI has a consortium called Daybreak for its cybersecurity model.
* A White House official stated that the administration does not provide approvals for private company AI releases.
* Engagements with government experts are voluntary, and decisions on release timing rest with the companies.
* The administration blocked Claude Mythos 5 and Fable 5 due to "national security concerns."
* OpenAI stated it would limit new models to "trusted partners" when complying with government requests.
* Chinese startup Moonshot AI unveiled its Kimi K3 model.
* David Sacks called the Kimi breakthrough "concerning," suggesting it impacts the AI race.
* The administration issued an executive order asking companies to voluntarily provide early access to models for testing.
* The administration launched a program called "Gold Eagle" to collaborate with the private sector on cyber vulnerabilities.
Executive Summary
The Trump administration is asserting greater control over the release of future artificial intelligence models by determining which companies and entities gain access to the latest frontier models. This contrasts with the previous system where American AI giants, such as Anthropic and OpenAI, controlled access to their most powerful models, often including major enterprise customers. While a White House official stated that the administration does not grant approvals for private company AI releases, they indicated a commitment to collaborating with frontier labs to enhance security without stifling innovation, noting that decisions on release timing rest with the companies.
Despite this stance, the administration has taken actions affecting model access, including blocking Claude Mythos 5 and Fable 5 due to national security concerns before reinstating access after negotiations. OpenAI also indicated a willingness to limit new AI models to "trusted partners" in response to government requests. This regulatory push occurs amid growing concerns about cybersecurity risks posed by sophisticated AI tools and the increasing competitive threat from open-weight models developed by entities like China's Moonshot AI. Furthermore, the administration has launched initiatives like the "Gold Eagle" program, intended to act as a clearinghouse for greenlighting access to new AI models.
Full Take
The dynamic described illustrates a tension between technological acceleration and state-centric control over foundational technologies. The transition from a market-led distribution model to one involving government oversight shifts the locus of power concerning AI safety and deployment. This shift forces a confrontation between commercial imperatives—driven by innovation and market competition, as seen in the race with open-weight models like Kimi—and national security objectives. The emergence of formal mechanisms like "Gold Eagle" suggests a pattern where perceived existential risk translates into regulatory intervention aimed at standardizing access and mitigating security risks.
The fact that previous voluntary arrangements (like Project Glasswing) are now being subject to explicit government clearinghouse approval indicates a systemic change in the relationship between private developers and governance structures. The concern raised by figures like David Sacks about losing the "AI race" suggests an underlying pattern where technological advantage is framed as a strategic national asset, prompting moves that prioritize security control over unfettered development speed. The system is attempting to manage emergent, high-stakes technology by inserting centralized gatekeeping functions into decentralized innovation streams.
The implication for human agency centers on whether necessary safety measures can be implemented without stifling the competitive drive that fuels innovation. If access to frontier models becomes conditional on government approval, it creates a complex dependency where innovation risk is potentially managed through centralized mandates rather than purely market forces. The critical question becomes whether this process successfully balances the imperative to secure technology against the potential for regulatory capture or stagnation of groundbreaking research. What external pressures—beyond immediate national security—are shaping these governmental frameworks, and who ultimately defines the acceptable balance between security and velocity in an AI-driven landscape?
Sentinel — Human
This text functions as standard journalistic reporting, synthesizing factual updates and attributed statements regarding the intersection of AI development and government oversight, showing typical features of professional news writing.