Trump calls for federal AI standard, warns China will ‘easily catch’ US

President Donald Trump on Tuesday called on Congress to establish a single federal standard for AI regulation, as House Republicans explore attaching preemption language to the National Defense Authorization Act that could override state AI laws nationwide.

In two posts on Truth Social, Trump urged lawmakers to act quickly. “Investment in AI is helping to make the U.S. Economy the ‘HOTTEST’ in the World,” he wrote. “But overregulation by the States is threatening to undermine this Major Growth Engine.”

The president warned that without unified standards, “China will easily catch us in the AI race. Put it in the NDAA, or pass a separate Bill, and nobody will ever be able to compete with America.”

Trump’s post aligns with efforts already underway in Congress. House Majority Leader Steve Scalise is exploring adding AI preemption language to the NDAA, a must-pass defense bill expected to be finalized in December. If successful, it would mark the most significant federal intervention in AI governance to date.

The cost of 50 rulebooks

The frustration behind Trump’s call stems from a regulatory landscape that has grown increasingly complex. State lawmakers introduced nearly 700 AI-related bills across 45 states in 2024, with 113 enacted into law.

Companies operating across state lines now face varying obligations under Colorado’s AI Act (effective June 2026), California’s SB 53 transparency requirements, and Texas’s TRAIGA (effective January 2026)—each with different requirements for bias testing, impact assessments, and consumer notification.

“The current patchwork of state-level AI rules has shifted from a background nuisance to a structural drag on execution,” said Sanchit Vir Gogia, chief analyst and CEO at Greyhound Research. “Enterprises building nationwide AI services now operate inside a maze where the same model can be considered high risk in one state, lightly constrained in another, and entirely unregulated at the federal level.”

Teams spend increasing time producing impact assessments, audit trails, and contract variants for different jurisdictions—costs felt most acutely by IT services providers embedding these constraints in multi-tenant platforms. Compounding the problem, a compliant model today may require redesign next quarter as new state mandates land.

The ‘woke AI’ trigger

Trump also pointed to ideological concerns as a driver for federal action. “Some States are even trying to embed DEI ideology into AI models, producing ‘Woke AI’ (Remember Black George Washington?),” he wrote, adding that federal standards could “protect children AND prevent censorship.”

The reference points to the February 2024 controversy when Google’s Gemini image generator produced historically inaccurate depictions, including Black Founding Fathers and racially diverse Nazi soldiers. Google paused the tool and CEO Sundar Pichai apologized, calling the outputs “problematic.”

The administration acted on these concerns in July with an executive order barring federal agencies from procuring AI models that fail “truth-seeking” and “ideological neutrality” standards.

The China argument, and its limits

Beyond ideological concerns, Trump framed federal preemption as essential to national competitiveness. His argument echoed industry positions. OpenAI, Google, and Nvidia CEO Jensen Huang have all backed federal preemption, arguing China’s centralized approach gives Beijing an advantage. Anthropic, however, has opposed broad preemption, lobbying against a proposed ten-year moratorium on state AI laws earlier this year.

The geopolitical picture is more complex than the rhetoric suggests, Gogia noted. China benefits from a single national rulebook, but AI leadership depends on capital, compute, talent, and the trustworthiness of systems being deployed.

“The argument that the US will be overtaken by China unless it replaces state rules with one federal AI standard is powerful rhetoric but analytically incomplete,” Gogia said. “The real competition will be won not just through fast deployment but through the ability to show that powerful AI systems are safe, resilient, and aligned with democratic values.”

Uncertain path, clear advice

Whether Congress acts on Trump’s call remains uncertain. Federal preemption faces headwinds: the Senate voted 99-1 in July to strip a moratorium provision from budget legislation. Forty state attorneys general opposed that measure, with California Attorney General Rob Bonta arguing that “states must be able to protect their residents by responding to emerging and evolving AI technology.”

Even if preemption passes, a weak federal standard won’t eliminate compliance challenges. Gogia said enterprises would still hold themselves to higher internal standards if federal rules underplay safety, transparency, and liability.

His advice to CIOs: don’t wait for Washington. The most resilient approach is to adopt the strictest credible regime as a baseline — typically EU AI Act obligations — and build modular systems that are adjustable by region, invest in transparency and documentation as rigorously as model accuracy.

“CIOs who pause major AI initiatives waiting for a perfect federal framework are exposing their organisations to far greater strategic risk than those who proceed under today’s rules with strong internal governance,” Gogia said.

The White House, OpenAI, and Anthropic did not immediately respond to requests for comment.