US Defends AI Frontiers Amid Accusations of ‘Industrial-Scale’ IP Theft by China

US agencies accuse Chinese AI firms of large-scale IP theft to accelerate market entry. A joint advisory warns of China's use of 'distillation' processes to extract US AI secrets. Tensions rise as plans to discuss AI theft surface ahead of a meeting between Donald Trump and Xi Jinping.

In a striking new development, United States federal agencies have leveled charges against leading Chinese artificial intelligence companies, accusing them of 'industrial-scale' theft of American trade secrets. This comes as a concerted effort to slash development costs and speed up their market entry, according to a joint advisory issued this Tuesday by the FBI, NSA, and CISA.

While some argue that the use of distillation techniques, where smaller AI models are trained on more complex architectures, can constitute legitimate research, US authorities and tech leaders insist that the scale and specificity of China's efforts classify them as blatant theft. The advisory adds weight to these accusations as geopolitical tensions around AI continue to mount.

President Donald Trump has announced that AI will be on the agenda when he meets with Chinese President Xi Jinping later this month in Washington, DC. The advisory highlights how Chinese AI firms, using a network of proxies dubbed 'transfer stations', bypass US regulations and security measures.

Companies like Alibaba Group and DeepSeek have been specifically named for conducting 'high-volume knowledge distillation campaigns' on US firms. US Treasury Secretary Scott Bessent has hinted at potential sanctions and the possibility of placing these firms on the Entity List.

Bessent also took to X (formerly Twitter) to assert that covert, industrial-scale actions crossing into IP theft could lead to severe consequences. Additionally, AI firm Anthropic reported an organized Chinese campaign attempting to exploit capabilities from its model, Claude, involving millions of interactions through fraudulent accounts.

The global scramble for AI superiority is raising cybersecurity red flags for American companies. A fresh report from Google's Threat Intelligence Group noted a worrying trend of hackers, motivated by various factors, targeting AI assets, foreshadowing a future where AI vulnerabilities are increasingly exploited for espionage and extortion.