BY EMMANUEL OGBONNA
The administration of Donald Trump has signaled a more aggressive stance against what it describes as the exploitation of American artificial intelligence technologies by foreign actors, with officials pointing in particular to entities in China amid intensifying global competition in the sector.
In a policy memorandum released Thursday, Michael Kratsios, the president’s chief science and technology adviser, outlined concerns that foreign organizations are systematically extracting capabilities from advanced U.S.-developed AI systems. The document alleges that such efforts—often referred to as “distillation”—amount to large-scale attempts to replicate the functionality of leading American models while bypassing the substantial costs and research investments required to build them independently.
Kratsios stated that the administration intends to work closely with domestic AI developers to detect these activities, strengthen technical safeguards, and explore punitive measures against those found to be engaging in unauthorized extraction. The proposed response underscores growing anxiety within Washington over maintaining technological leadership in artificial intelligence, a field increasingly viewed as central to economic competitiveness and national security.
The policy push comes at a time when China is rapidly advancing its capabilities, narrowing what was once a significant gap with the United States. A recent assessment by Stanford Institute for Human-Centered Artificial Intelligence concluded that performance differences between top-tier AI models from the two countries have effectively diminished, intensifying the strategic stakes.
Chinese officials have rejected the U.S. accusations, describing them as unfounded and politically motivated. Liu Pengyu, representing Beijing’s embassy in Washington, said China remains committed to technological progress through fair competition and international cooperation, while emphasizing its support for intellectual property protections.
Similar objections were voiced in Beijing, where Guo Jiakun dismissed the claims as baseless and accused the United States of attempting to undermine China’s technological achievements. He called on Washington to abandon what he characterized as prejudiced views and instead foster greater collaboration in scientific research.
The issue has also gained traction in the U.S. Congress. The House Foreign Affairs Committee recently expressed unanimous bipartisan support for legislation designed to identify and penalize foreign actors accused of extracting key technical features from proprietary American AI systems. Proposed measures include sanctions and other restrictions targeting entities found to be involved in such practices.
Supporters of the bill argue that model extraction represents a new frontier in economic competition and intellectual property disputes. Bill Huizenga, who sponsored the legislation, warned that safeguarding AI innovation is critical as these technologies become increasingly influential in areas ranging from cybersecurity to military applications.
Tensions escalated further following the emergence of DeepSeek, a Chinese startup that drew global attention after unveiling a large language model capable of rivaling leading U.S. systems at significantly lower cost. The development unsettled markets and raised questions about how quickly advanced AI capabilities can be replicated.
David Sacks, who previously advised the administration on artificial intelligence and cryptocurrency, suggested that such breakthroughs may have relied on distillation techniques involving American models. He pointed to what he described as evidence that knowledge from systems developed by