China expands attacks to steal U.S. AI and targets startups

The race for leadership in artificial intelligence is increasing pressure on technology companies in the United States. Experts in security point out that groups linked to China have expanded their espionage strategies, no longer seeking only specific technological projects but also exploiting human and operational vulnerabilities at companies that develop AI.

According to Matt Pearl, director of the strategic technologies program at the Center for Strategic and International Studies, the intensification of competition between the two countries has led Chinese agents to broaden their interest in the artificial intelligence sector.

“With the intensification of the AI race, the [People's Republic of China] has increasingly targeted the technology sector,” he said.

According to Pearl, the objective is not limited to the theft of a single trade secret. Information on product planning, supply chains, commercial strategies, and internal processes has come to be considered valuable for reducing the technological gap between Chinese and American companies.

Recent reports reinforce this concern. In June, CrowdStrike reported that entities linked to China accounted for more than half of state-sponsored intrusions against technology companies during the 12 months ending in March, with a special focus on assets related to artificial intelligence.

Anthropic also accused Chinese companies, including Alibaba, of trying to illegally obtain information about its AI technologies. The Chinese company did not comment on the allegations.

Another episode drew attention after Copyleaks released an analysis indicating that responses produced by DeepSeek's R1 model showed similarities with content generated by ChatGPT in approximately three quarters of the tests conducted.

“We did not see [the same stylistic match] in other law master's degrees,” said Alon Yamin, CEO and co-founder of Copyleaks.

In addition to attempts to obtain technology, experts warn of the growth of operations seeking to place professionals inside American startups. Brian Abbott, founder of Agentiq Capital, said that a hired employee deliberately altered the company's code and content before being dismissed. The case was referred to the FBI, although the accusation has not yet been independently verified.

“China's economic espionage campaign is a constant threat that costs the American economy hundreds of billions of dollars per year and puts our national security at risk,” the FBI said.

Experts emphasize that startups are among the main targets because they normally have small digital security teams. According to Cliff Steinhauer, of the National Cybersecurity Alliance, artificial intelligence-driven social engineering campaigns have made employees one of the most vulnerable points of companies.

“We have seen many cases within our company of new employees who, upon joining the company, immediately become targets of cyberattacks to gain access to our AI models,” Yamin said.

Isaac Stone Fish, founder of the consultancy Strategy Risks, believes that the actions attributed to Beijing have grown significantly since the race for artificial intelligence gained momentum. In his view, in addition to cyberattacks, supply chain restrictions, pressure on employees, and government incentives are part of the technological competition between China and the United States.