Trump Administration’s AI Brain Trust Navigates Complex China Policy Amid Deep Divisions

As China’s open-weight Artificial Intelligence models demonstrate increasingly powerful capabilities, the Trump administration finds itself at a critical crossroads, entrusting a small, disparate group of officials with the weighty task of determining the scope and nature of restrictions to be imposed. This influential cadre, scattered across various departments and agencies throughout Washington, is united by the challenge but deeply divided on the optimal strategy to ensure the United States maintains its lead in the accelerating global AI race against China.

The internal discord within the administration is significant, reflecting a complex issue with multifaceted implications. "It’s not an argument with two sides, it’s an argument with ten sides," one senior White House official candidly told WIRED, underscoring the profound lack of interagency coordination. In such an environment, policy decisions are almost inevitably shaped by those who command the most direct influence with the president. To shed light on the intricate forces at play in this high-stakes regulatory debate, a closer look at the key power brokers involved reveals their individual stances and departmental mandates.

Commerce Secretary Howard Lutnick

Howard Lutnick, the Commerce Secretary, has emerged as a particularly influential figure in the ongoing intra-administration debate concerning AI policy. His department’s Bureau of Industry and Security (BIS) is responsible for overseeing export controls, granting him a powerful lever in shaping technology transfer policies. Lutnick’s approach appears to occupy a middle ground within the broader regulatory spectrum.

Sources familiar with the discussions indicate that Lutnick has actively explored various mechanisms to incentivize leading US AI laboratories to develop and release their own open-weight models. The strategic aim here is clear: to create a robust domestic counterweight to China’s burgeoning AI ecosystem. In pursuit of this objective, he has engaged in direct conversations with leaders from several prominent AI research labs, seeking their input and cooperation.

Lutnick’s cautious assessment of Chinese AI capabilities was evident in a recent X post where he downplayed the performance of Moonshot AI’s Kimi K3 model. He stated that his team’s benchmarks indicated Kimi K3 performed worse than leading US frontier models in critical tests, suggesting a desire to temper concerns about China’s immediate lead. While exhibiting a more freewheeling stance than some hardliners within the White House, Lutnick has demonstrated a willingness to impose controls when deemed necessary. This was exemplified by his department’s imposition of export controls on Anthropic, effectively bringing the company into compliance with certain administrative expectations regarding its powerful models.

Arvind Raman, Acting Director, CAISI

Working closely under Secretary Lutnick, Arvind Raman has assumed a pivotal role as the acting chief at the Center for AI Standards and Innovation (CAISI). Located within the Commerce Department, CAISI serves as the primary conduit for communication and collaboration between the burgeoning AI industry and the US government. Raman’s appointment to this critical position occurred just last week, following the abrupt resignation of his predecessor, Chris Fall.

Fall’s departure was reportedly linked to weeks of intensive negotiations with Anthropic, as he pressed the company to implement more robust safeguards to prevent "jailbreaks" – instances where users circumvent protective measures – in its most advanced Fable 5 model. WIRED previously reported that CAISI staff engaged in numerous meetings with Anthropic’s technical teams, working diligently to establish new safeguards deemed sufficient to allow Fable 5 to be brought back online. Raman now inherits the responsibility of continuing CAISI’s vital work in setting and upholding AI standards, particularly as it relates to national security and competitive advantage.

National Cyber Director Sean Cairncross

Taking a decidedly more hardline stance, National Cyber Director Sean Cairncross has emerged as a key advocate for stringent measures against Chinese AI labs. His position grants him an outsized influence over policies aimed at countering potential national security risks posed by advanced AI. Cairncross played a crucial role in drafting President Donald Trump’s June 2 executive order, which established a comprehensive framework for assessing the most powerful AI models, signaling the administration’s serious intent to regulate this emerging technology.

Despite a professional background rooted in political campaign law, most recently with the Republican National Committee, and a confessed lack of prior tech or AI experience, Cairncross has earned credit from both current and former staff. They commend his earnest efforts to grapple with the complex national security challenges presented by AI and his leadership style, which allows his staff significant autonomy without micromanagement.

A particular focus for Cairncross and his chief of staff, Lara Smith, has been the practice of "distillation," where Chinese labs train their models by leveraging or "distilling" the outputs of superior US models. This practice is seen as a direct threat to American intellectual property and technological superiority. White House spokesperson Elizabeth Huston affirmed the administration’s overall strategy, stating, "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," emphasizing a focus on bolstering US AI capabilities.

Former AI Czar David Sacks

Even after departing his official role as AI czar in March, tech investor David Sacks has maintained a significant and influential advisory relationship with President Trump on matters of Artificial Intelligence. Sacks consistently advocates for a hands-off, laissez-faire approach to AI regulation across the board. His influence was notably demonstrated when he successfully intervened at the eleventh hour to temper some of the more restrictive regulatory provisions initially proposed in the June 2 executive order.

Sacks extends his consistent stance to Chinese open-weight AI models, leveraging his considerable presence on X (formerly Twitter), where he commands 1.6 million followers, to publicly influence administration policy from the outside. In a July 23 X post that garnered nearly 500,000 views, Sacks publicly supported Secretary Lutnick’s assessment, writing, "Secretary Howard Lutnick is right. The Kimi Panic needs to stop. American frontier models are still ahead. When you factor in what’s in the lab, the gap is even larger. As long as we keep releasing, we will stay ahead," articulating a belief that continuous innovation and open release of US models is the best defense against foreign competition.

White House Chief of Staff Susie Wiles

As President Trump’s chief political aide, Susie Wiles holds an indispensable position in the AI policy formation process. She serves as the crucial gatekeeper, meticulously vetting and scrutinizing all policy proposals from every side of the debate before they are presented to the Oval Office. Given the fragmented nature of the administration’s approach to AI, Wiles’ role in consolidating and prioritizing these diverse viewpoints is paramount.

High-ranking officials, including Treasury Secretary Scott Bessent, Commerce Secretary Howard Lutnick, and National Cyber Director Sean Cairncross, routinely present their departmental positions directly to Wiles. Her decisions, much like on other complex administration issues, frequently prove to be the final word, effectively shaping the policy agenda that ultimately reaches the president’s desk.

Treasury Secretary Scott Bessent

Scott Bessent, the top Trump official responsible for US-China trade relations, has adopted arguably the most aggressive stance within the administration regarding Chinese AI and the controversial practice of distilling US models. His department’s increasing involvement signals a broader concern about the economic and strategic implications of China’s technological advancements.

In a recent X post, Bessent unequivocally condemned distillation as "IP theft," drawing a clear line in the sand. He went further, threatening Chinese labs engaged in such practices with severe repercussions, including sanctions or inclusion on the Entity List—a powerful trade-restriction roster maintained by the US government that effectively bars listed entities from acquiring certain US technologies.

Initially playing a minimal role in the discussions surrounding restrictions on Anthropic last month, Bessent has significantly escalated his involvement in AI policy. This heightened focus comes ahead of key diplomatic engagements, including Chinese President Xi Jinping’s anticipated visit to the White House in September and the APEC summit in November, where US-China economic and technological competition is expected to be a central theme.

According to a source familiar with his thinking, Bessent has communicated to Wiles and other senior officials the urgent need for the US, Europe, Japan, and other Western allies to accelerate the development of their own open-weight AI models. He firmly believes that failure to do so risks ceding the critical AI race to China. While Bessent directly engages with Wiles at the highest levels, his top deputy, Luke Pettit, plays a crucial role in executing this aggressive roadmap at the staff level.

Assistant Secretary for Financial Institutions at the Treasury, Luke Pettit

Under the leadership of Secretary Scott Bessent, Luke Pettit, the Assistant Secretary for Financial Institutions at the Treasury Department, has become the department’s dedicated point person on AI policy. Pettit, who previously served as a senior policy adviser in the US Senate, now works in tandem with the Treasury’s chief information officer and former DOGE member Sam Corcos, bringing a blend of legislative experience and technological understanding to the fore.

The Treasury Department, under Bessent and Pettit, is particularly focused on the catastrophic potential of "rogue" AI. Sources indicate a deep concern about the possibility of an AI model inadvertently triggering economic instability, for instance, by autonomously closing out trading positions or illicitly accessing sensitive information held by major financial institutions like Goldman Sachs. Such an event, they contend, could precipitate a severe economic crisis. This focus highlights a unique dimension of the Treasury’s involvement, emphasizing the profound economic and systemic risks posed by unchecked or insecure AI advancements.

The ongoing deliberations within the Trump administration underscore a pivotal moment in US technology policy, where national security, economic competitiveness, and the future of AI innovation hang in the balance. With diverse opinions and approaches, the final policy framework regarding Chinese AI models will be a testament to the influence wielded by these key figures.