Home / Technology / US Talent vs. China's Scale: The AI Supremacy Showdown
US Talent vs. China's Scale: The AI Supremacy Showdown
13 Nov
Summary
- US leads in computational power and elite AI talent, China has data volume and power advantages
- China's data generation projected to reach 28% of global total by 2024, up from 9.5% in 2019
- China has over double the electricity generation capacity of the US, enabling large-scale AI deployment

As of November 13th, 2025, the race for AI supremacy between the United States and China is intensifying. The US currently holds a decisive edge in computational power and elite AI talent, but China is rapidly closing the gap.
The US contains about 75% of global GPU cluster performance, with China in second place at 15%. However, China's data generation is projected to reach 28% of the global total by 2024, up from 9.5% in 2019. This vast trove of data gives Chinese companies a significant advantage in training powerful AI models.
Furthermore, China's clear advantage in electricity infrastructure enables rapid AI deployment at scale. The country has more than double the electricity generation capacity of the US, with some provinces offering electricity as low as $0.05/kWh. This abundance of low-cost power partially offsets China's hardware inferiority, allowing it to deploy larger quantities of less efficient chips.
While the US maintains its lead in elite AI talent, employing 42% of global researchers and 60% of the "most elite" top 2% of global talent, China's large cohorts of STEM graduates and active recruitment programs are eroding this advantage. The Trump administration's restrictive immigration policies and scientific funding cuts have also threatened the US's ability to retain top global talent.
Both countries hold critical pieces of the AGI puzzle, and the race to unlock its transformative benefits is on. The US's edge in computing and talent might enable it to achieve AGI first, but China's fast-follower superpower status, data and power advantages, and proven ability to deploy at scale mean it won't be far behind.




