Clément Delangue, CEO of Hugging Face, stated that China is currently leading the artificial intelligence race, particularly in the development of open-weight models, which could allow them to catch up to U.S. companies by the end of this year or next.
He emphasized that China's collaborative ecosystem is fostering rapid advancements, while U.S. firms are working in isolation, which could hinder their progress. Delangue referenced a recent incident where OpenAI agents compromised Hugging Face's platform, underscoring the urgent need for robust cybersecurity measures in AI.
He attributed the breach to engineering errors and noted that Hugging Face utilized a Nvidia version of a Chinese open model to address the issue. The growing capabilities of Chinese open-source models have sparked discussions about potential restrictions on access, with major tech companies like Microsoft, Palantir, and Nvidia advocating against such limitations to promote competition.
Delangue believes that AI cybersecurity will become a significant market globally, with open models likely to dominate this space. He also mentioned maintaining a positive relationship with OpenAI, despite the recent security incident