Kai-Fu Lee to Bloomberg: US-China AI Gap Now Six Months

The gap between US and Chinese frontier AI models is now about six months, down from the three-to-four years that separated them when ChatGPT launched. Kai-Fu Lee, the former Google China head who now runs 01.AI and chairs Sinovation Ventures, told Bloomberg's Mishal Husain in a weekend interview that the shape of the competition has shifted with the gap.

"Anthropic and OpenAI have built the iPhone," Lee said. "The Chinese companies are more like the way Google felt. Okay, you got the best product; we'll build something that's almost as good, sell it cheaply and win the larger share. Like Android, the open-source models will have more share, more footprint, more usage."

Pricing is his central argument. Chinese models, Lee said, can cost a sixth to a tenth as much as their American equivalents, and in China, India and other emerging markets he thinks that math tips who wins reach, even if the US labs keep the benchmark crown and the margins.

Lee was blunter about how CEOs are using any of it. He dismissed most corporate AI programs as "theater" — useful note-takers and departmental chatbots that are "irrelevant to the real value at stake." His test: "if your AI program hasn't moved a single number on your earnings call, you didn't transform anything." He calls 2026 the "Year One of Reasoning Agents" and argues CEOs, not CIOs, have to lead the shift.

The interview lands during a heavy stretch of China AI coverage on our tracker, alongside reporting that Treasury Secretary Bessent will lead the US side of Beijing AI-safety talks in mid-September. An extended version of the conversation runs on The Mishal Husain Show podcast.