From TIMEWELL
This is Hamamoto from TIMEWELL.
Looking for AI training and consulting?
Learn about WARP training programs and consulting services in our materials.
An Intense Global Competition
In the 21st century, the rapidly evolving field of artificial intelligence has become the arena for fierce global competition. Among the competing nations, the US and China stand out — displaying notable differences in investment levels, the number of AI models developed, research paper output, and patent filings, yet continuing to push each other forward. Stanford University's AI Index Report captures the numbers, strategies, and even the atmosphere of the development environment in comprehensive detail. This article draws on that report to analyze the US-China AI development competition across multiple dimensions: investment levels, the performance of top AI models, and trends in papers and patents.
The US maintains a lead in developing many of the major AI models, backed by overwhelming investment and an impressive track record. China, on the other hand, is rapidly gaining ground in the quantitative dimensions — the volume of publications and patent filings — with universities and research institutions in particular delivering results that are clearly reflected in the numbers. Recent trends also suggest that the performance gap between top AI models from each country is narrowing, indicating that the two powers are moving into increasingly direct technological competition.
These dynamics reflect each country's distinct strategy. The US pursues a high-level development approach centered on private enterprises, maintaining top-tier technology through a secretive approach. China pursues what might be called a "second-mover strategy," focusing on patent filings and paper publications to aim for leadership in international technology standards. In simple terms: the US is a "big investment, winner-takes-all" approach; China is a "share knowledge broadly, catch up from behind together" team approach.
This article first examines the current state of investment and major AI model development, then analyzes how the differences in papers and patents reflect the divergent strategies of each country. It also examines the strategic significance of these data points and their implications for the future of AI technology development.
- The US Leads on Investment — and China Is Closing the Gap
- The US-China Strategy Through AI Papers and Patents — China Leads in Volume; the US Leads in Quality and Rule-Setting
- The Strategic Core That Separates US and Chinese AI Development — Secrecy as the #1 Player, Openness as the Challenger
- Summary
The US Leads on Investment — and China Is Closing the Gap
Investment is one of the most fundamental factors driving AI innovation. Modern AI development demands enormous funding, and where that money flows significantly shapes the future of the technology. According to Stanford University's AI Index Report, US private investment in AI in 2024 reached approximately $100 billion — roughly 15 trillion yen — while China's stood at approximately $9.3 billion, or about 1.5 trillion yen. By the numbers alone, the US is pouring more than 12 times as much money into AI development as China.
Behind the US's ability to mobilize such investment is the presence of a domestic ecosystem of giant tech companies. Firms like OpenAI, Google, and Microsoft bring enormous research budgets and talent to the development of cutting-edge AI models on a daily basis. Beyond direct technology development, these companies also focus on disseminating research results through patents and papers, as well as participating in international rule-making — a broad-based approach that is paying dividends.
Also worth noting is the US's substantial lead in the number of major AI models. The report lists 40 notable AI models attributable to the US versus 15 for China — a clear indication of both quantitative and qualitative dominance in US technology development. Summarizing the key points:
- The US leads China by a wide margin in private investment
- The US also leads in the number of major AI models
- The US maintains technology through a company-driven, high-level, and largely secretive approach
These points demonstrate that US advantage extends beyond financial firepower to encompass technological capability and organizational strength. With the backing of major companies investing heavily even in generative AI, US private AI investment in that segment alone reached $29 billion — exceeding the combined investment of China, Europe, and the UK. This suggests the US is not merely numerically superior but strategically ahead as well.
China, for its part, is trailing on investment but pursuing technology development through its own methods and strategies. China's approach is characterized by a heavy emphasis on rapid growth in quantitative research output — publications from universities and public research institutes, along with patent filings. Though it lags the US significantly in private investment scale, the breadth and vigor of its overall research activity and its diverse approaches in generative AI in particular are beginning to attract attention.
On the performance dimension, there was once a significant gap between the top AI models from the US and China — but in recent years, that gap has been gradually narrowing. Metrics from venues like Chatbot Arena show signs that Chinese models are catching up to what was previously an American-dominated ranking, suggesting that competition will intensify further going forward.
In terms of investment and the number of AI models developed, the US currently maintains an overwhelming lead. But the narrowing performance gap between top models is a highly significant signal for thinking about how technological innovation will unfold going forward. The competitive landscape is becoming increasingly challenging.
Beyond technology development, capital flows also play a major role in creating new markets and establishing international standards. In areas like generative AI, autonomous driving, and robotics, the US's commanding financial strength often translates directly into international competitiveness — putting pressure on research institutions in Japan and Europe to strengthen their own standing as well.
The US-China Strategy Through AI Papers and Patents — China Leads in Volume; the US Leads in Quality and Rule-Setting
In AI development, the volume of published papers, citation counts, and patent filings serve as important indicators of each country's technology strategy and research maturity.
China's strength is its rapid and accelerating publication output. Since 2016, the volume of AI research papers has risen steadily, with researchers — particularly from academia — actively publishing results of the latest technologies. China's papers account for 22.6% of global citations, a figure that reflects growth momentum not fully captured by raw numbers alone. But the more telling comparison emerges when looking specifically at the top 100 most impactful papers.
Geographically, the top 100 papers break down as 50 from the US and 34 from China. For high-quality research — work excelling in international recognition and citation counts — the US continues to hold a clear advantage. This reflects the fact that US companies like Google, Microsoft, and NVIDIA demonstrate high research capability not only as private enterprises but in academia as well, and their output contributes to international technology standards and rule-making in practice.
Also notable is that in AI theory, safety, and responsible AI, the US publishes more than twice as many papers as China. In responsible AI specifically, the US has 669 papers to China's 268 — positioning the US to lead international discourse in this area. As AI's practical deployment advances, the ethical and social dimensions of responsible AI will only grow in importance, making this a strategically critical domain.
It's also worth examining the distinction between papers and patents as output metrics. Papers primarily serve to share knowledge and receive academic evaluation — their number and citation count reflect the breadth and influence of research activity. Patents, on the other hand, are a critical tool for protecting newly developed technology and creating commercial value. In patent filings, the data is striking: approximately 70% of all AI-related patents filed worldwide in 2023 came from China. From 2014 to 2023, about 6,300 patents were registered in the US, while China registered approximately 38,000 — roughly six times more.
This surge in patent filings reflects China's strategic effort to protect its technical output and gain a stronger voice in international standardization forums — even as it has not yet fully caught up to the US in top model performance.
Key points on papers and patents:
- China leads the world in total paper volume, with citation counts on the rise — but in the highest-impact top 100 papers, the US holds more than half.
- In patent filings, China leads the US by several multiples — reflecting an aggressive stance on technology protection and international standardization.
- In AI ethics and safety — including responsible AI — the US publishes research far outpacing China.
Looking at both papers and patents, the approaches of the two countries diverge clearly. The US, driven by companies, pursues high quality in technology and delivers strong results in international rule-making and academic evaluation. China, through research volume expansion via universities and public institutions and technology protection through patent filing, is pursuing a bold strategy designed with its own long-term interests in mind.
These data points illuminate a complex reality that can't be reduced to a simple "which is stronger" binary. As AI technology continues to advance and new markets are created, how each approach functions in international competition — and how the differences shape global technology standards — will remain an important theme to watch. The high-quality research of the US and the rapid quantitative growth of China, as they intersect, are likely to lead AI's future toward a more dynamic and diverse trajectory.
The Strategic Core That Separates US and Chinese AI Development — Secrecy as the #1 Player, Openness as the Challenger
The differences between US and Chinese AI development go beyond investment levels and numbers of papers and patents — they extend to the underlying strategy itself. The US has firmly established its position as the "first mover," building an advanced R&D environment with a strong emphasis on quality, trade secret protection, and confidentiality. For top-tier AI models, major US companies tend to minimize public disclosure of technical details and closely guard them as trade secrets — avoiding the risk of competitors easily catching up and aiming to maintain a dominant position in cutting-edge technology over the long term.
China, meanwhile, is steadily advancing a strategy designed to challenge the US as the "second mover." Chinese companies and research institutions, even if not yet fully matching the performance of top US models, are working to accumulate quantitative proof of their technology and know-how — building international influence and negotiating leverage. In concrete terms, the rapid growth in patent filings and the volume of paper publications are designed to gain a stronger voice at international conferences and in technology standardization processes.
These strategic differences are clearly reflected in both corporate management and national-level policy. US companies like OpenAI, Google, and Microsoft are pouring enormous investment into maintaining their top position on every front while adhering to a secretive management approach — holding technical core elements within the company rather than quickly making research public, thereby protecting competitive advantage. By contrast, Chinese companies tend to share technical information openly, promoting ecosystem building with a broad base of users, researchers, and international partners.
Corporate development strategy also extends beyond pursuing top model functionality to the formation of the ecosystem the technology creates. By releasing open-weight models, users can freely use and customize them, ultimately accelerating both market adoption and technology diffusion. This kind of strategy — a flexible Chinese approach — stands in contrast to the American model, which involves higher upfront investment and higher technical barriers.
The two countries' strategies also exert significant influence on international rule-making and standardization debates. The US is working to drive rule-making in responsible AI and safety discussions, backed by its commanding volume of papers, aiming to embed its own technical standards in global platforms. China, backed by its surge in patent filings, is moving to speak more actively in international standardization forums and shape future technology standards to its advantage.
The strategic divergence will have direct implications for future technology trends and market formation. The American approach, backed by the strength of leading companies in maintaining quality in cutting-edge technology, is likely to continue delivering high-performance, highly reliable products and services. The Chinese approach, through large-scale paper publishing, patent filing, and open information sharing, is expected to generate new value in international competition through rapid ecosystem expansion.
Looking ahead, there is a possibility that the US and China will each leverage their respective strengths to create a form of mutual complementarity in international AI standardization. For example, US companies providing high-security, high-performance technology while Chinese companies and institutions lead ecosystem building and discussions at international forums — this kind of division of labor may contribute to shaping the global AI market going forward.
The strategic differences between the "first mover" and the "second mover" fundamentally define the nature of each country's AI development competition, and intense back-and-forth is expected to continue in both technological innovation and market competition. Researchers, companies, and policymakers worldwide cannot afford to look away from the outcomes these strategies produce — and how they ultimately affect our lives and businesses.
Drawing on Stanford University's AI Index Report and detailed transcripts, this article has analyzed the US-China AI development competition from multiple angles. In investment levels and the number of major AI models, the US maintains an overwhelming lead backed by extraordinary financial firepower and a strong track record. At the same time, the narrowing performance gap between top AI models signals that technological competition is likely to intensify further.
In paper publication volume, China leads the overall count — but the US still holds a majority of the highest-impact top 100 papers, reflecting the quality advantage. In patent filings, China has recorded roughly six times the number of the US over the past decade, clearly manifesting a strategy of protecting research output and pursuing international standardization.
The AI development strategies of both countries extend beyond mere numerical comparison to the underlying approaches, corporate cultures, and consciousness of international rule-making. The US drives top-tier quality while leveraging massive investment and a commitment to confidentiality. China, as the challenger, is rapidly expanding its technology ecosystem and building its voice in international forums through paper publishing and patent filing — aiming for the leading role in future international technology standards.
The strategic differences will significantly affect the future expansion of the AI technology market, the intensification of international competition, and the building of global ecosystems. The "first mover" strategy of the US and the "second mover" strategy of China may complement each other — ultimately leading to improvements in international competitiveness through technological innovation. For example, US companies delivering high-quality products and services while China takes responsibility for ecosystem building for market penetration — a future in which the strengths of both converge is worth anticipating.
The numbers and discussions raised in this article are not merely data points — they are important clues for forecasting the future direction of AI development. For companies, research institutions, and international regulatory bodies, these shifts in the competitive landscape are themes that demand close attention — with direct relevance to our lives and business environments. How the US and China will continue to spark new strategies and technological innovations going forward is something we cannot afford to stop watching.
In closing, this article reinforces that AI has already become the arena for intense competition that transcends national borders. Financial power, technical capability, and output metrics like papers and patents are the critical keys to global competition going forward — and understanding them is an indispensable element for anyone seeking to read the future of technology trends. The US-China AI race symbolizes the future we are all moving toward, and as a valuable example of both its intensity and its possibility, it will continue to attract attention for years to come.
Reference: https://www.youtube.com/watch?v=5LhXdivbeiM