Jorge Costa Oliveira
The Stanford HAI AI Index Report 2026 was recently released – a document that deserves close attention. One finding stands out: “the number of AI researchers and programmers moving to the U.S. has fallen 89% since 2017, with an 80% drop in the past year alone.” While the U.S. still hosts more AI talent than any other country, it is attracting new talent at the lowest rate in over a decade.
What explains such a sharp decline?
Several factors identified by Stanford HAI, MacroPolo, and CSET help explain why foreign AI talent is no longer choosing the United States in the same numbers.
First, cost of living and market saturation. The soaring expense of cities like San Francisco and Seattle, combined with tech layoffs in 2023 and 2024, has weakened the appeal of the “American dream.” Many developers now see emerging markets or their home countries as more stable and offering better quality of life.
Second, the decentralization of AI. It is no longer necessary to be in Silicon Valley to stay at the frontier. Cities such as Toronto, London, Paris, Tel Aviv, and Lisbon have become competitive hubs, often with more welcoming immigration policies. Remote work also allows top engineers to work for U.S. firms without relocating.
Third, geopolitics and national security tensions. The climate between the U.S. and China – including past initiatives targeting Chinese researchers and ongoing export controls – has created an atmosphere of suspicion. Many Chinese scientists feel constrained or scrutinized, discouraging relocation and pushing some to return home.
Fourth, China’s rise as a talent powerhouse. Data shows China now produces a larger share of top AI researchers. Around 40% of leading AI experts in top U.S. labs are Chinese, while NVIDIA CEO Jensen Huang has suggested the figure could be closer to 50%. Unlike in 2017, when most Chinese PhDs stayed in the U.S., a growing number now return home, drawn by strong investment and competitive salaries.
Fifth, immigration policy and visa barriers. This remains the most pressing issue. The U.S. system is widely seen as outdated for a digital economy. AI post-graduate programs depend heavily on international students, yet tightening visa rules are reducing enrollment. Green card backlogs for nationals from countries such as China and India can last decades, pushing talent toward Canada or the U.K. Meanwhile, uncertainty around H-1B visas has intensified, with costs rising to about $100,000 per hire.
Some tech leaders, including Sam Altman and Jensen Huang, support stricter rules, arguing they filter for elite talent. But the broader impact is clear. Hardline immigration policies are weakening the pipeline that sustained U.S. technological leadership.
This trend is especially evident among Chinese specialists, but also applies to Indian talent. Faced with visa hurdles and geopolitical tension, many are staying home or turning to Europe.
AI is not just a race of capital – it is, first and foremost, a race for talent. The trillions invested by companies such as Microsoft, Meta, and Google assume a steady supply of top researchers – an assumption that is increasingly fragile.
Given AI’s importance to the U.S. economy, draconian restrictive immigration policies risk serious consequences. A shrinking talent base could erode competitiveness and ultimately cost the United States its leadership in AI innovation and top products.
No Comments