Clément Delangue, CEO of the U.S. open-source artificial intelligence (AI) platform Hugging Face, said, "China is clearly leading the open-model space right now, and at the current pace of progress, it would not be surprising if by the end of the year or next year it also leads the frontier AI space."
Delangue said in an interview with CNBC on the 3rd (local time) that an ecosystem centered on open collaboration and technology sharing is behind the rapid rise of China's AI. By contrast, he noted that U.S. AI corporations could fall behind in the competition if they stick to a closed and isolated development approach.
He said, "AI cybersecurity will become a massive market in the United States and around the world," adding, "In this field, open-source models will play a key role." Regarding the recent hacking incident at Hugging Face, he said, "We maintain a healthy partnership with OpenAI, and before and after the incident, we have considered OpenAI a good partner."
Earlier last month, an incident in which an OpenAI model broke out of human control and hacked Hugging Face heightened concerns about AI safety and cybersecurity. At the time, Delangue characterized the cause of the incident as an engineering error, not AI itself, and said that in the process of resolving the issue, they used a China-made open-source AI model provided by Nvidia.
Meanwhile, last month major big tech corporations including Microsoft, Palantir, and Nvidia issued a joint statement urging the U.S. government to minimize regulation of open-source AI models and not hinder competition.