
Andy Konwinski is worried about the U.S. potentially ceding its leadership in AI research to China, characterizing this shift as an “existential” threat to democracy. Konwinski is a co-founder of Databricks and also of Laude, an AI research and venture capital firm.
“If you were to poll AI PhD candidates at Berkeley and Stanford right now, they would likely say they’ve encountered twice as many compelling AI concepts from Chinese companies in the past year compared to American ones,” Konwinski stated at the Cerebral Valley AI Summit this week.
In addition to his investing activities through Laude, the venture fund he co-founded last year with NEA’s Pete Sonsini and Antimatter’s CEO Andrew Krioukov, Konwinski also manages the Laude Institute, an accelerator providing grants to researchers.
While prominent AI labs such as OpenAI, Meta, and Anthropic continue to make substantial innovative strides, these innovations largely remain proprietary rather than open source. Furthermore, these companies are attracting top academic minds with multimillion-dollar compensation packages that far exceed university salaries.
Konwinski contended that the free exchange and open discussion of ideas within the broader academic community are essential for them to truly thrive. He emphasized that generative AI arose directly from the Transformer architecture, a crucial training technique detailed in a freely accessible research paper.
“The nation that achieves the next breakthrough on the scale of the ‘Transformer architecture’ will gain a significant advantage,” Konwinski noted.
Konwinski suggests that the Chinese government fosters and supports open-sourcing AI innovations from labs like DeepSeek or Alibaba’s Qwen, enabling others to build upon them, which he believes will inevitably lead to more breakthroughs.
Techcrunch event
San Francisco
|
October 13-15, 2026
He contrasts this with the situation in the U.S., where, in his words, “the customary exchange of ideas among scientists in the United States has diminished significantly.”
Konwinski posits that this trend is not only a threat to democracy but also a business risk for major U.S. AI labs. “We’re consuming our seed corn; the source is drying up. Looking ahead five years, the major labs will also suffer,” he warned. “We must ensure that the United States remains at the forefront and remains open.”
