China Accused of Stealing American AI Secrets, FBI Launches Investigation

The artificial intelligence rivalry between the United States and China is intensifying, with hackers from Beijing reportedly attempting to narrow the technological gap by focusing on stealing American secrets. Matt Pearl, director of the strategic technologies programme at the Center for Strategic and International Studies, stated that as the AI competition heats up, the People’s Republic of China is increasingly targeting the technology sector. He noted that China is no longer just focusing on specific trade secrets but is broadening its efforts to understand a company’s product roadmap, particularly in competitive sectors, and to identify weaknesses in supply chains. According to Pearl, this strategy is a way to close the AI gap with the US by three to four months.

Several cases of alleged theft have recently come to light. In June, American cybersecurity giant Crowdstrike claimed that entities linked to China were responsible for more than half of the attacks targeting technology companies, especially those in the AI field, over the previous 12 months. AI company Anthropic also levelled similar accusations against several Chinese firms, including Alibaba, alleging they attempted to steal its AI capabilities. Furthermore, US-based AI startup Copyleaks previously claimed that DeepSeek’s R1 model produced responses nearly identical to ChatGPT, suggesting it may have been trained on American AI models. Copyleaks CEO and co-founder Alon Yamin remarked that they had not seen such stylistic similarities with other large language models.

The FBI has described Chinese espionage campaigns as a national security threat to the United States. The bureau confirmed it is investigating the risk of technology theft by foreign parties and is committed to protecting the homeland.