US Losing AI Dominance to China, Warns Databricks Co-Founder

According to Databricks co-founder Andy Konwinski, the US is facing an "existential" threat as it loses its dominance in AI research to China. Konwinski, who also co-founded the AI research and venture capital firm Laude, says that PhD students at top US universities like Berkeley and Stanford are now reading twice as many interesting AI ideas from Chinese companies compared to American ones.

Konwinski argues that for AI innovations to truly flourish, they need to be freely exchanged and discussed within the larger academic community. He points to the Transformer architecture, a pivotal training technique introduced in an open-source research paper, as an example of how open-source breakthroughs can lead to major advancements. In contrast, Konwinski says the "diffusion of scientists talking to scientists" that has historically driven US innovation has "dried up" as major AI labs like OpenAI, Meta, and Anthropic keep their work largely proprietary.

The Laude Institute founder believes the Chinese government's support and encouragement for open-sourcing AI innovations from labs like DeepSeek and Alibaba's Qwen will inevitably lead to more breakthroughs. He warns that this trend poses a risk not only to democracy, but also to the long-term success of the US AI industry, as "the fountain is drying up" and the country risks "eating its corn seeds."