Ein tiefer Blick in das Talent und die Auswirkungen von Deepseek Ai für uns Innovation

“Ultimately, DeepSeek AI represents more than just another advance in language model technology. It reveals talent patterns that challenge long-held US assumptions about innovation advantage.” [**Amy Zegart**](https://www.hoover.org/profiles/amy-zegart) and coauthor **Emerson Johnston** offer this other findings in a new paper analyzing the human talent responsible for DeepSeek’s R1 language model and V3 general-purpose large language model (LLM). As the authors write, „Nearly all of the researchers behind DeepSeek’s five papers were educated or trained in China. More than half of them never left China for schooling or work, demonstrating the country’s growing capacity to develop world-class AI talent through an entirely domestic pipeline.“ The authors suggest that this „robust pipeline of homegrown talent“ within China should unsettle American assumptions about remaining an unparalleled destination for high-skill workers, particularly in technology, in years ahead. The paper also explores in detail the US institutional affiliations and international relocation patterns of DeepSeek researchers, shedding new light on the model maker’s global talent pool.