US Security Experts Warn AI Companies Aren’t Protecting Secrets Enough

Google and AI Security Concerns

Formation of Security Committee

Google,​ along with⁢ its associated​ services and apps like ChatGPT, recently established a⁢ security committee on its board. This move highlights growing concerns about security vulnerabilities in the AI sector.

US Export Controls and Chinese AI Development

The US has imposed export controls that hinder Chinese developers ‍from creating‍ their ‍own⁣ AI models. According to Matheny, this has led to an increased risk of cyberattacks⁤ aimed at stealing AI software. ⁤He noted that spending a few million dollars on such cyberattacks is a worthwhile investment for China, given ⁣that developing these‌ models could cost American companies⁣ hundreds of billions of dollars.

“It’s really hard, and it’s really important, and we’re not investing enough nationally to get that right,” Matheny said.

Response from China

China’s embassy in Washington, DC, ‌did⁣ not ​respond to⁤ WIRED’s⁤ request for comment on these theft accusations. Historically, they have dismissed such‌ claims as baseless smears by⁣ Western officials.

Incident of AI Chip Secrets Theft

Google reported an incident to law enforcement that⁢ led to a US case alleging the theft of AI chip secrets for China. Despite Google’s stringent safeguards, it ⁣took considerable time to catch‌ the accused, Linwei Ding, a Chinese national‍ who has pleaded not guilty to federal charges.

Details of the Alleged‍ Theft

Prosecutors allege that‌ Linwei ‍Ding, also known as Leon,​ was hired in 2019 to ‍work⁤ on software for Google’s supercomputing data centers. Starting in 2022, ⁢he allegedly copied over 500 files containing confidential information to his personal Google account. The scheme involved pasting information into Apple’s Notes app on his company ⁤laptop, converting the files to PDFs, and uploading ​them elsewhere, all while evading ⁤Google’s security measures.

Potential Consequences

During the alleged theft, the US claims Ding was in contact with the⁣ CEO of a Chinese AI startup and was planning to start his own AI company in China. If convicted, he faces up to 10 years in prison.