In a comprehensive report published in WIRED magazine by senior political correspondent Hugo Lowell on July 27, 2026, the complex state of shaping artificial intelligence (AI) policy in the administration of United States President Donald Trump is laid bare. As China’s open-weight AI models grow increasingly powerful, the Trump administration relies on a small, decentralized group of officials and administration members to decide whether and what restrictions should be imposed on the technology. This group, scattered across several different departments and agencies in Washington, presents a wide range of approaches and opinions on the best way to maintain the United States' position as a leader in the race against China. "It’s not an argument with two sides, it’s an argument with ten sides," a senior White House official told WIRED in describing the current situation within the administration.
Due to the lack of structured and ongoing coordination between different agencies in the administration, policy is expected to ultimately be set by the figures with the greatest influence on the President himself. The report presents a detailed mapping of the key forces involved in the race to regulate and govern artificial intelligence in the current administration.
Commerce Secretary Howard Lutnick: The Mediator Between Restrictions and Freedom of Action
US Commerce Secretary Howard Lutnick is gradually emerging as an influential and central figure in the intra-administration dispute regarding artificial intelligence, partly because his department oversees export controls through the Bureau of Industry and Security (BIS). According to sources familiar with the matter, Lutnick has contemplated ways to create incentives for leading AI labs in the United States to develop their own open-weight models, aiming to serve as a counterbalance to China. As part of these efforts, he has spoken with leaders of several leading AI labs.
Last Thursday, Lutnick played down the capabilities of the Chinese model Kimi K3, developed by Moonshot AI. In a post on the social network X (formerly Twitter), Lutnick wrote that his team found the Chinese model performed worse than leading US frontier models in key benchmark tests. Lutnick appears to be straddling a middle ground on regulation. On one hand, he imposed export controls on Anthropic to bring the company in line with administration requirements, but on the other hand, he demonstrates a more freewheeling approach compared to other figures in the White House.
Arvind Raman: Acting Director of the Center for AI Standards and Innovation
Arvind Raman has become Lutnick’s top deputy as the acting director of the Center for AI Standards and Innovation (CAISI). The center, which operates under the Commerce Department, serves as the industry’s primary point of contact with the federal government. Raman assumed his role just last week, after his predecessor, Chris Fall, abruptly resigned.
Fall’s resignation came after he spent weeks trying to persuade Anthropic to add stronger safeguards and security mechanisms to prevent "jailbreaks" of its most powerful model, Fable 5. CAISI staff members held lengthy meetings with Anthropic’s technical teams for weeks as both sides tried to hammer out new safeguards deemed strong enough to allow the Fable 5 model to be brought back online, as previously reported in WIRED magazine.
National Cyber Director Sean Cairncross: The Hardline Stance Against China
National Cyber Director Sean Cairncross takes a much tougher and more rigid approach toward Chinese AI labs. Cairncross plays a particularly significant role in potential attempts to regulate Chinese AI, and he was empowered by the White House to develop policy to counter potential national security risks arising from this technology. He helped draft Donald Trump’s June 2 executive order, which established a framework for assessing the most powerful AI models.
Cairncross, a former political campaign lawyer who recently served at the Republican National Committee (RNC), comes to the role without prior experience in technology or AI. Despite this, current and former staff members credit him with trying to grapple with the complex national security problems posed by AI, while allowing his staff to work without micromanaging every problem in the way a software engineer might.
In particular, Cairncross and his chief of staff, Lara Smith, are focused on curbing Chinese labs from developing and training their models based on leading US models, a practice known as "distillation." In response to a request for comment, White House spokesperson Elizabeth Huston stated that the administration’s deliberations on Chinese AI are centered on trying to bolster and reinforce US AI. "The United States leads the world in AI innovation, and President Trump will keep it that way. The Trump administration is doubling down on innovation to widen the gap between America and the rest of the world," Huston said.
Former AI Czar David Sacks: The Free-Market Voice
Tech investor David Sacks remains one of Trump’s most influential advisers on AI, maintaining a direct line to the president even after leaving his official role as the administration's AI czar in March. Sacks is an ardent supporter of a complete hands-off approach regarding AI, and he even successfully intervened at the last minute to water down some of the regulatory provisions included in the June 2 executive order.
Sacks has remained consistent in his laissez-faire approach toward Chinese open-weight models as well. He uses his X account, which has about 1.6 million followers, to influence the administration from the outside. "Secretary Howard Lutnick is right. The Kimi Panic needs to stop. American frontier models are still ahead. When you factor in what’s currently in the lab, the gap is even larger. As long as we keep releasing, we will stay ahead," Sacks wrote in an X post on July 23 that received nearly half a million views.
White House Chief of Staff Susie Wiles: The Gatekeeper of Presidential Decisions
As Trump’s top political aide in this process, White House Chief of Staff Susie Wiles carefully vets and analyzes policy proposals coming from all different sides and camps before they reach the President's desk in the Oval Office. Treasury Secretary Scott Bessent presents his department’s positions directly to Wiles, as do Lutnick and Cairncross. Similar to other issues in the administration, Wiles' decisions are ultimately the ones that end up being final and decisive.
Treasury Secretary Scott Bessent and Deputy Luke Pettit: Anxiety Over Economic Stability
As the highest-ranking official in the Trump administration in charge of trade relations between the United States and China, Treasury Secretary Scott Bessent has adopted perhaps the most aggressive stance toward Chinese AI and China's attempts to perform "distillation" of US models. In a post on X last week, Bessent rejected and condemned the practice of distillation as "IP theft" and threatened Chinese labs with sanctions or inclusion on the "Entity List"—the US government's official trade-restriction list.
After playing virtually no role in imposing restrictions on Anthropic last month, Bessent significantly stepped up his involvement in AI policy ahead of Chinese President Xi Jinping's expected visit to the White House in September and the planned APEC summit in November. Bessent has told Wiles and others that the US, Europe, Japan, and other Western allies must move faster to develop their own open-weight models, or else risk losing the AI race to China, according to a source close to his thinking.
While Bessent works directly with Wiles at the White House, one of his top deputies, Luke Pettit—Assistant Secretary for Financial Institutions at the Treasury—is involved in implementing and executing the roadmap formulated by Bessent at a professional staff level, according to the same source. Pettit, who previously served as a senior policy adviser in the US Senate, is now working alongside the Treasury’s Chief Information Officer (CIO) and former DOGE member, Sam Corcos.
The entire Treasury Department has been focused in recent months on economic risks and the possibility of AI models going rogue ("going rogue"). According to internal sources, the greatest fear is that an AI model could, even inadvertently, close out trading positions in financial markets or obtain access to sensitive information at leading banks such as Goldman Sachs, thereby causing a total collapse of the American economy.