The US leads China in AI computing power, holding about 75% of global capacity versus China’s 14%, with Nvidia supplying over 60% of global AI chips. China is closing the gap in research output and models, while a proposed US‑China AI notification mechanism signals growing national‑security and compliance pressure for businesses.
What Happened
On 24 September 2026, United States President Donald Trump and China’s President Xi Jinping were slated to meet in Washington, DC, for a summit covering trade, artificial intelligence (AI), Taiwan and the US‑Israel war on Iran. Ahead of the talks, US Treasury Secretary Scott Bessent announced that Washington had proposed an AI “notification mechanism” with China – a hotline to alert each other when AI incidents threaten national security.
Al Jazeera’s visual explainer compares the US and China’s AI strength across four dimensions: computing power, AI models, spending and research. Training and running AI requires enormous amounts of computing power. The more powerful chips and computers a country has, the faster it can build and improve AI models, making computing power one of the biggest advantages in the AI race. Computing power is measured in FLOP/s (floating‑point operations per second) – or how many calculations a computer can perform in one second. Adding up the computing power of all the AI chips a country has gives a sense of its total AI capacity. By that measure, the US is well ahead, accounting for nearly three‑quarters of the world’s total AI computing power, while China holds just more than 14 percent, according to Epoch AI, an AI research institute. The US lead comes largely from its access to the most advanced chips. US company Nvidia accounts for more than 60 percent of global AI computing capacity among major chip designers, while China’s Huawei holds a much smaller share.
In AI models, the US dominates with a larger share of state‑of‑the‑art models, but China is closing the gap. China’s leading research institutions have published several advanced models in recent months, and Chinese firms are investing heavily in next‑generation architectures. In AI research, China leads in publication output and citation impact, while the US remains ahead in overall research funding and patent filings.
What This Means For You
As an AI practitioner or business leader, the US‑China AI race has immediate implications for talent acquisition, supply chain resilience and regulatory compliance. First, the computing power gap suggests that US‑based AI firms will continue to have faster iteration cycles, giving them a competitive edge in developing cutting‑edge models. If your organization relies on high‑performance GPUs, consider diversifying suppliers to mitigate geopolitical risks. Nvidia’s dominance means that any US‑China trade tensions could disrupt your hardware pipeline; explore alternative chip makers or cloud GPU options from European or Asian vendors.
Second, the growing Chinese research output signals that Chinese firms may soon offer comparable or superior models for specific domains, such as natural language processing in Mandarin or image recognition for industrial applications. Keep an eye on Chinese model releases and evaluate their performance against your current solutions. If you operate in multilingual markets, integrating Chinese models could unlock new customer segments.
Third, the proposed AI notification mechanism reflects a growing focus on AI safety and national security. If your organization develops or deploys AI that could impact critical infrastructure, be ready to comply with emerging reporting requirements. Engage with industry groups to shape standards that balance innovation with risk mitigation.
Finally, the broader geopolitical context means that AI policy will likely tighten. Stay informed about US‑China AI trade rules, export controls on advanced chips, and potential sanctions on AI technologies. Align your procurement and R&D strategies with these evolving regulations to avoid costly compliance breaches.
Why It Matters
This suggests that the AI advantage is shifting from sheer computational capacity to a more nuanced balance of talent, research output, and investment. While the US remains the leader in overall computing power, China’s rapid gains in model sophistication and research output could erode that lead in specific application areas. The proposed notification mechanism underscores the growing recognition that AI incidents can have national‑security implications, prompting a move toward more coordinated risk management between the two superpowers.
Key Takeaway
- US holds ~75 % of global AI computing power; China ~14 %.
- Nvidia supplies >60 % of global AI chips; Huawei’s share is much smaller.
- China’s AI research output is rapidly closing the gap with the US.
- US‑China summit includes an AI “notification mechanism” for national‑security incidents.
Leave a Reply