Chinese National Accused of Stealing AI Secrets From Google

A Chinese citizen who recently quit his job as a software engineer for Google in California has been charged with trying to transfer artificial intelligence technology to a Beijing-based company that paid him secretly, according to a federal indictment unsealed on Wednesday.

Prosecutors accused Linwei Ding, who was part of the team that designs and maintains Google’s vast A.I. supercomputer data system, of stealing information about the “architecture and functionality” of the system, and of pilfering software used to “orchestrate” supercomputers “at the cutting edge of machine learning and A.I. technology.”

From May 2022 to May 2023, Mr. Ding, also known as Leon, uploaded 500 files, many containing trade secrets, from his Google-issued laptop to the cloud by using a multistep scheme that allowed him to “evade immediate detection,” according to the U.S. attorney’s office for the Northern District of California.

Mr. Ding was arrested on Wednesday morning at his home in Newark, Calif., not far from Google’s sprawling main campus in Mountain View, officials said.

Starting in June 2022, Mr. Ding was paid $14,800 per month — plus a bonus and company stock — by a China-based technology company, without telling his supervisors at Google, according to the indictment. He is also accused of working with another company in China.

Mr. Ding openly sought funding for a new A.I. start-up company he had incorporated at an investor conference in Beijing in November, boasting that “we have experience with Google’s 10,000-card computational power platform; we just need to replicate and upgrade it,” prosecutors said in the indictment, which was unsealed in San Francisco federal court.

“The Justice Department will not tolerate the theft of artificial intelligence and other advanced technologies that could put our national security at risk,” said Attorney General Merrick B. Garland, who announced the indictment during an appearance at an American Bar Association conference in San Francisco on Wednesday afternoon.

The charges underscore the high-stakes contest for primacy in artificial intelligence. While American companies have developed most advances in generative A.I., China has made it a strategic priority to lead the growing field.

Tech industry insiders have estimated that China is at least a year behind the United States, but many Chinese start-ups have tapped American technology to try to keep up, especially Meta’s open-source large language model, called Llama. Generative A.I., which is behind ChatGPT and the wave of conversational chatbots, has quickly become one of the world’s most coveted technologies.

In seconds, these types of tools can generate convincing text and images that could be used to boost productivity, create misinformation or provide amusement. Audio and video capabilities are not far behind. Google developed some of the foundational breakthroughs that make these systems work. The company has said that its latest group of A.I. models, named Gemini, are among the most powerful available today.

But since ChatGPT’s debut, Google has lost its status as a market leader and its stumbles have attracted attention. The company has been widely criticized for racial biases in its image generator, leading it to pause users’ ability to create images of people.

Accusations of intellectual property theft have been a major sticking point in U.S.-China relations for years. A Chinese national was arrested in 2015 for selling some of IBM’s source code to parties in China. In 2018, a former Apple employee was apprehended as he tried to board a flight to Beijing with the company’s autonomous-driving trade secrets.

The same year, the Chinese firm Sinovel Wind Group was convicted of stealing wind turbine technology from a Massachusetts-based company, AMSC, which incurred more than $800 million in losses.

In October, Christopher A. Wray, the F.B.I. director, said that intellectual property theft from China was a danger to U.S. economic and national security, describing it as the “defining threat of this generation.”

José Castañeda, a Google spokesman, said in a statement: “We have strict safeguards to prevent the theft of our confidential commercial information and trade secrets. 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 F.B.I. for helping protect our information and will continue cooperating with them closely.”

The indictment suggested that Mr. Ding had some help, saying that another Google employee swiped Mr. Ding’s identification card at a company office to help him conceal a trip to China.

Google, referring to Mr. Ding as a “junior employee,” initially said he had acted alone but later said that did not appear to be the case. It maintained that its security systems had worked as intended.

It was not immediately clear whether Mr. Ding has legal representation.

The government offered few details about the life of Mr. Ding, who began working for Google in early 2019 and quit suddenly in January — after booking a one-way ticket to Beijing.

Mr. Ding listed a degree from the Dalian Institute of Technology in China in 2010, along with degrees from the University of Southern California and Stanford, on a LinkedIn page that corresponded to his name and the details of employment at Google.

The page lists stints at software semiconductor and health care companies over the past decade, along with awards he said he earned at Google, including the “Perfy Award and Feats of Engineering.”

Kitty Bennett contributed reporting.