U.S. Businesses Urge Trump Administration Not to Restrict Access to Chinese Open-Source AI Models

Business owners who increasingly rely on open-source AI models are urging the Trump administration not to restrict access to Chinese models. China, which has rejected accusations that it copied the Frontier Model, is also calling on the United States to listen to the business community—or risk falling further behind economically.

U.S. AI policy may be focused on the wrong layer

A policy research report conducted for Congress and submitted to the U.S.-China Economic and Security Review Commission in March suggested that the United States may already be at a disadvantage by focusing too heavily on maintaining its lead in frontier AI while failing to build a strong domestic open-source ecosystem.

China recognized this gap and moved to fill it, according to the researchers.

“U.S. export controls are tailored to limit frontier training by restricting access to advanced semiconductors. They do not support small-scale model deployment cycles that do not require very high computing power, utilize openly available base models, and generate benefits through application rather than pre-training. If the most important models for industrial AI are small, specialized, and open, the current U.S. policy framework may be targeting the wrong layer of competition.”

China’s open-source AI strategy raises economic and security concerns

Companies including Nvidia and Meta have pledged to help the United States catch up. However, the researchers warn that losing the country’s extensive AI user base to China could weaken the U.S. ability to set technical standards and norms for AI development “for years to come.”

Some members of Congress are also concerned that a loss of global AI leadership could damage U.S. national security and the economy.

Congress weighs restrictions on Chinese AI models

U.S. Rep. John Moolenaar (R-Mich.), chairman of the House China Committee, took a tougher position, insisting that “no federal agency should be using Chinese AI models,” according to Reuters.

“Doing so would only increase the federal government’s dependence on Chinese AI models, which is not in the national interest,” Moolenaar said.