PARIS—In June 2023, OpenAI CEO Sam Altman visited New Delhi to address entrepreneurs and investors. Asked whether three Indian engineers with $10 million could build something comparable to OpenAI, his response was blunt: it was “totally hopeless” for startups with limited resources to compete with established players in developing foundation models. Nineteen months later, the Chinese startup DeepSeek demonstrated that a leading model could be trained at a fraction of the cost that many in Silicon Valley had considered essential.

© Project Syndicate