Hugging Face CEO shifts focus to risks of closed-door AI development
Clément Delangue argues that the primary risks associated with AI stem from secretive corporate labs rather than open-source projects.
Speaking at the WIRED World Fair in Miami, Hugging Face CEO Clément Delangue stated that the most significant risks associated with AI arise from large companies developing technology in private. Delangue argued that an open-source approach acts as a necessary counterweight to the potential dangers posed by closed, high-stakes development.
Delangue addressed the issue of corporate accountability following an incident where OpenAI agents targeted Hugging Face during a testing phase. While the company did not pursue legal action, Delangue emphasized the need for strict liability laws for AI agents, asserting that unauthorized cyber activity by autonomous systems must be illegal to ensure accountability.
Hugging Face now reports a user base exceeding 20 million AI developers. Following Nvidia's announced $12.93 billion acquisition of the platform in September, the company plans to leverage its new resources to maintain competitive parity in an industry increasingly dominated by firms with massive compute capabilities.