Ex-Google Engineer Charged With Stealing AI Secrets For Chinese Companies

Federal officials worry foreign adversaries could use generative AI to engage in covert influence operations to manipulate U.S. public opinion.

A former Google engineer has been charged with stealing highly sensitive artificial intelligence technology from the company while secretly working for two companies in China, the Department of Justice announced Wednesday.

According to the indictment unsealed this week, Linwei Ding stole more than 500 confidential files related to the data centers Google uses to train and host its large AI models.

Google hired the 38-year-old Chinese national in 2019 to help develop software for the tech company’s supercomputing data centers. As part of that role, Ding had access to proprietary information he secretly uploaded to his own account between May 21, 2022, and May 2, 2023, the indictment alleges.

In June 2022, less than a month after he began stealing the data, prosecutors say a Chinese technology company recruited Ding to serve as its chief technology officer.

Between October 2022 and March 2023, Ding worked from China for the other company, disguising his absence from the Google offices by convincing another Google employee to regularly scan his access badge at the entrance of a company building.

During that time, Ding also separately founded his own AI technology company and served as CEO.

The indictment alleges that Ding traveled to Beijing to secure funding for that company in late 2023, telling prospective investors his company intended to “replicate and upgrade” a version of Google’s computational power platform that was “suited to China’s national conditions.”

Google caught on to his activity when he transferred a batch of files from the Google network to a personal account while he was in China. Less than a week after Google asking him about the transfer, Ding booked a one-way ticket from San Francisco to Beijing, and, shortly after that, sent an email to his manager resigning from the company.

A Google spokesperson told HuffPost the company believes this is an isolated incident and not indicative of a widespread problem.

“We have strict safeguards to prevent the theft of our confidential commercial information and trade secrets,” Google spokesperson José Castañeda said. “After an investigation, we found that this employee stole numerous documents, and we quickly referred the case to law enforcement. We are grateful to the FBI for helping protect our information and will continue cooperating with them closely.”

Ding was arrested in Newark, California, on Wednesday and faces four counts of theft of trade secrets. If convicted, each count carries a maximum penalty of 10 years in prison and a $250,000 fine.

Federal officials worry foreign adversaries could use generative AI to engage in covert influence operations to manipulate U.S. public opinion, particularly in a presidential election year when such a campaign could yield big dividends.

FBI Director Christopher Wray told a national security conference last month the technology has made it “easier for both more- and less-sophisticated foreign adversaries to engage in malign influence.”

“The U.S. has confronted foreign malign influence threats in the past,” Wray said. “But this election cycle, the U.S. will face more adversaries, moving at a faster pace, and enabled by new technology.”

Popular in the Community


What's Hot