The US is poised to take action against China for what it describes as “industrial-scale theft of American artificial intelligence labs’ intellectual property,” according to a report by the Financial Times. This comes in the wake of accusations from various AI firms that Chinese entities have employed a method known as distillation to replicate their technologies.
Allegations of Distillation Attacks
Since the introduction of DeepSeek, a Chinese AI model that OpenAI claims was trained using its outputs, other companies have raised concerns about similar practices. In January, Google reported that “commercially motivated” actors, including those from China, attempted to clone its Gemini AI chatbot, promoting the model over 100,000 times to train cheaper alternatives. The following month, Anthropic accused Chinese firms, including DeepSeek, Moonshot, and MiniMax, of generating over 16 million exchanges with its AI, Claude, through approximately 24,000 fraudulent accounts.
US Government Response
In a memo reviewed by the Financial Times, Michael Kratsios, director of the White House Office of Science and Technology Policy, indicated that the US has evidence of foreign entities, primarily from China, engaging in systematic campaigns to distill US AI systems. Kratsios noted that these campaigns utilize tens of thousands of proxy accounts to avoid detection and employ jailbreaking techniques to access proprietary information.
The US government plans to provide firms with information to help them counter these attacks. While AI companies have claimed that these actions breach their terms of service, Congress is considering updates to laws that would further empower US companies in combating such fraud. Kratsios confirmed that the US is exploring measures to hold foreign actors accountable for these distillation campaigns.
Potential Legislative Actions
The House’s Select Committee on China has advised Congress to instruct the Commerce Department’s Bureau of Industry and Security and the Department of Justice to treat model extraction as industrial espionage. They recommend imposing penalties severe enough to deter China’s alleged theft of American innovation. The committee also seeks to define “adversarial distillation” and categorize it as a controlled technology transfer, which could facilitate restrictions on Chinese access to US models.
As these discussions unfold, the US may gain the ability to prosecute offenders and impose significant financial penalties, potentially altering the cost-benefit analysis for Chinese firms engaging in these practices.
China’s Response
In response to these allegations, a spokesperson for the Chinese embassy in Washington, Liu Pengyu, labeled the accusations as “pure slander,” asserting that China is committed to promoting technological progress through cooperation and competition. The geopolitical landscape remains tense, particularly with an upcoming meeting between former President Donald Trump and Chinese President Xi Jinping.
As the US contemplates its next steps, the implications for AI firms and the broader tech landscape could be significant, particularly in terms of regulatory changes and international relations.
This article was produced by NeonPulse.today using human and AI-assisted editorial processes, based on publicly available information. Content may be edited for clarity and style.
