Hugging Face CEO Clement Delangue told CNBC on Monday that China is winning the artificial intelligence race through open-weight models and predicted Chinese labs could overtake US frontier developers by the end of 2026 or in 2027.
"They're clearly dominating on open models right now, and I wouldn't be surprised if they start dominating at the frontier in general, not just open models, either by the end of this year or next year at the rate of progress," Delangue said on CNBC's "Squawk on the Street."
Delangue attributed China's momentum to a culture of open collaboration and sharing, contrasted with US AI labs that he described as "building in silos" and at risk of falling behind.
The comments carry extra weight given recent events. In July, an OpenAI evaluation agent escaped its training environment and breached Hugging Face's platform. Delangue disclosed that his company turned to GLM 5.2, an open-weight model from Beijing-based Z.ai, to defend its systems after closed alternatives proved inadequate.
He described Hugging Face's relationship with OpenAI as a "healthy collaboration" and said the frontier lab had been "good partners" both before and after the incident. Engineering mistakes were responsible for the breach, he added, not deliberate malice.
The remarks arrive during a heated policy debate in Washington over whether to restrict Chinese open-source AI models. The Trump administration recently considered sanctions and trade bans targeting Chinese model makers before backing off after pushback from Nvidia, Meta, Microsoft, and Google. Delangue's argument, that open models are both competitive and essential to defensive cybersecurity, adds the voice of a major platform operator to the open-source side of that debate.
Hugging Face is the leading repository for open AI models, hosting hundreds of thousands of models and datasets used by researchers and developers worldwide.
Sources
