This interview podcast features Alexandr Wang, CEO of ScaleAI, discussing the AI race between the US and China. The conversation begins with Wang's surprising statistics revealing that leading Chinese AI models are on par with, or even surpass, their American counterparts, despite US export controls on advanced chips. He highlights the release of DeepSeek's model, which performed exceptionally well on a new benchmark test ("Humanities Last Exam"), contradicting the common assumption of US AI superiority. Wang then discusses the competitive landscape of large language models (LLMs), emphasizing the diversity of capabilities among different models and the increasing importance of data generation in pushing AI boundaries. Finally, the conversation touches upon the massive investment needed in AI infrastructure and the potential for the LLM market to grow to trillions of dollars in the next few years.
Sign in to continue reading, translating and more.
Continue