Garry Tan, CEO of Y Combinator, has expressed support for American AI labs using distillation techniques to learn from frontier AI models, a practice currently under scrutiny due to allegations against some Chinese labs. Distillation involves extensively querying a model to understand its reasoning and capabilities, which can then inform the training of new models. While some companies like Anthropic have accused Chinese labs of illicit distillation, Tan argues that US labs should be free to engage in similar practices openly and legally.
Tan believes that restricting how users interact with AI models through API calls is overly limiting and that government intervention could help establish norms treating access to AI intelligence as a public good rather than a proprietary asset. He points out that many proprietary AI models were trained on vast amounts of publicly available data without explicit permission from content owners, suggesting a precedent for open access.
He envisions a balanced AI ecosystem where frontier labs continue to innovate and receive funding, while open-weight models provide broader access and freedom for users. Tan warns against a scenario where AI power consolidates within a single dominant company, which could stifle competition and innovation. His stance highlights the ongoing debate about openness, intellectual property, and regulation in the rapidly evolving AI industry.