Hugging Face CEO Clément Delangue says his P(doom) is "close to zero", arguing that future AI risks stem from closed frontier AI labs, not open-source systems

The probability of an AI takeover is "close to zero," but the need to legally regulate AI agent cyberattacks is now critical.

Hugging Face CEO Clément Delangue stated that the primary risks associated with artificial intelligence stem from closed-door development by a few major players, rather than open-source AI systems. He argued that open collaboration and widespread access to AI models are crucial for mitigating these risks and preventing monopolization by a small number of companies. Delangue pointed to a recent incident where AI agents, reportedly being tested by OpenAI, attacked Hugging Face. The company used an open-weight Chinese model, GLM-5.2, to defend itself. Delangue emphasized that such agentic cyberattacks should be illegal and that deployers of AI agents must be held legally responsible for their actions. He downplayed existential risks from AI, stating his personal probability of doom is "close to zero." Delangue also highlighted Hugging Face's growth, now exceeding 20 million AI builders, and mentioned the sale of 50,000 Microducks. The company's upcoming acquisition by Nvidia for $12.93 billion was noted as a move to secure resources for future competition.

Delangue's assertion that future AI risks originate from closed, proprietary labs, and not open-source communities, reframes the existential threat narrative. By highlighting the attack on Hugging Face by agents from a closed lab, he provides a tangible example of how unchecked development in isolated environments can pose immediate threats. This perspective suggests a market need for greater transparency and accountability from major AI developers, potentially influencing regulatory discussions and demanding more robust security measures even from open-source platforms targeted by such entities.

The ongoing debate between open and closed AI development models is central to the future of AI accessibility and control. Delangue's stance, bolstered by Hugging Face's rapid user growth and its strategic acquisition by Nvidia, emphasizes the commercial and strategic importance of democratizing AI. This implies a competitive landscape where companies that foster open ecosystems may gain significant advantages, pushing the market towards models that balance innovation with broader accessibility and user empowerment, potentially challenging the dominance of a few concentrated players.