HuggingFace CEO Raises Concerns Over Chinese Open-Source AI Models and Censorship

HuggingFace CEO has concerns about Chinese open source AI models

However, they have also attracted criticism - including from OpenAI employees - for censoring topics sensitive to the Chinese government, such as the Tiananmen Square massacre. HuggingFace's CEO Clement Delangue says he has similar concerns. In a recent podcast (in French), he warned about the unintended consequences of Western companies building on top of well-performing, open source Chinese AI. "If you create a chatbot and ask it a question about Tiananmen, well, it's not going to respond to you the same way as if it was a system developed in France or the US," Delangue warned. Delangue noted that if a country like China "becomes by far the strongest on AI, they will be capable of spreading certain cultural aspects that perhaps the Western world wouldn't want to see spread." Delangue has previously stated that Chinese AI is quickly catching up to Western AI thanks to its embrace of the open source movement. Delangue warned on the podcast that the strong concentration of top open source models coming from China is a "fairly new development and I'm a little worried about it to be honest," he said. "It's important that AI is distributed between all countries - that there isn't one or two countries which are much stronger than the others." HuggingFace is the world's biggest platform for AI models and a popular place for Chinese AI companies to showcase their latest LLMs. In fact, HuggingFace's CTO announced this week that the default model on HuggingChat is Qwen2.5-72B-Instruct, which was developed by Chinese e-commerce giant Alibaba. This particular model does not appear to censor questions about the Tiananmen Square massacre or other issues typically censored by the Chinese government. A different model from Alibaba's Qwen family available on HuggingChat, QwQ-32B, however, clearly does when TC asked: DeepSeek, another Chinese model which went viral in the AI community for its reasoning capabilities, also extensively censors topics deemed sensitive by the Chinese government, TechCrunch previously reported. Chinese AI companies are in a tough spot as the Chinese government forces their models to "embody core socialist values" and comply with its already-extensive censorship system. A HuggingFace spokesperson declined further comment but pointed out that Delangue recently predicted China would start to lead the global AI race in 2025.

Western AI Firms Are Worried About 'Censorship' In China's AI Models

AI models in China have been gaining popularity as their capabilities continue to grow. As China's open-source AI models continue to grow in prevalence and catch up to their Western counterparts, some leading AI firms have shared concerns over censorship. By focusing on open-source development, companies such as Alibaba and Moonshot AI have made rapid advances in the AI space. Censorship and Culture In September, Clement Delaunge, the CEO of U.S. AI startup HuggingFace, warned that the growth of Chinese models could have far-reaching consequences for the Western world. "If you create a chatbot and ask it a question about Tiananmen, well, it's not going to respond to you the same way as if it was a system developed in France or the U.S.," Delaunge told French podcast Génération Do It Yourself. The CEO said that if China becomes the strongest on AI, it will be capable of "spreading certain cultural aspects that perhaps the Western world wouldn't want to see spread." HuggingFace, the leading platform for AI models, features hundreds of LLMs from Chinese AI companies. According to a report from TechCrunch, China-based DeepSeek, which gained popularity for its extensive reasoning capabilities, was found to censor topics historically sensitive to the Chinese government. Delaunge previously stated that China was getting close to catching up with AI in the West due to its embrace of open source. Other leading AI companies have spoken out against Chinese AI models, too. OpenAI developer Steven Heidel posted a meme on X relating to China's AI censorship of the Tiananmen Square massacre. In 2023, OpenAI CEO Sam Altman shared concerns about "authoritarian regimes" developing competing AI technology. "We do worry a lot about authoritarian governments developing [AI models]," Altman told ABC News . Popularity Is Increasing for China's AI Models China's AI models have been getting an increasing spotlight recently as their capabilities continue to grow. Chinese models have shown themselves to excel in bilingual benchmarks, which has helped to expand their global appeal. The country's ability to rapidly iterate on open-source Western frameworks has also contributed to this progress. Alibaba Cloud said its new set of AI models, QwQ, allegedly managed to outperform OpenAI's o1 in some benchmark tests. According to the tech company, the AI models did better in two maths tests while matching o1 in problem-solving and coding.