Global AI Race Intensifies: China Closes Gap with US as Competition Heats Up

AI race in 2025 is tighter than ever before

The artificial intelligence (AI) race is heating up: the number and quality of high-performing Chinese AI models is rising to challenge the US lead, and the performance edge between top models is shrinking, according to an annual state of the industry report. The report highlights that as AI continues to improve quickly, no one firm is pulling ahead. On the Chatbot Arena Leaderboard, which asks users to vote on the performance of various bots, the top-ranked model scored about 12% higher than the tenth-ranked model in early 2024, but only 5% higher in early 2025 (see 'All together now'). "The frontier is increasingly competitive -- and increasingly crowded," the report says. The Artificial Intelligence Index Report 2025 was released today by the Institute for Human Centered AI at Stanford University in California. The index shows that notable generative AI models are, on average, still getting bigger, by using more decision-making variables, more computing power and bigger training data sets. But developers are also proving that smaller, sleeker models are capable of great things. Thanks to better algorithms, a modern model can now match the performance that could be achieved by a model 100 times larger two years ago. "2024 was a breakthrough year for smaller AI models," the index says. Bart Selman, a computer scientist at Cornell University in Ithaca, New York, who was not involved in writing the Index report, says it's good to see relatively small, cheap efforts such as China's DeepSeek proving they can be competitive. "I'm predicting we'll see some individual teams with five people, two people, that come up with some new algorithmic ideas that will shake things up," he says. "Which is all good. We don't want the world just to be run by some big companies." The report shows that the vast majority of notable AI models are now developed by industry rather than academia: a reversal of the situation in the early 2000s, when neural nets and generative AI had not yet taken off. Industry produced fewer than 20% of notable AI models before 2006, but 60% of them in 2023 and nearly 90% in 2024, the report says. The United States continues to be the top producer of notable models, releasing 40 in 2024, compared with China's 15 and Europe's 3. But plenty of other regions are joining the race, including the Middle East, Latin America and southeast Asia. And the previous US lead in terms of model quality has disappeared, the report adds. China, which produces the most AI publications and patents, is now developing models that match their US competition in performance. In 2023, the leading Chinese models lagged behind the top US model by nearly 20 percentage points on the Massive Multitask Language Understanding test (MMLU), a common benchmark for large language models. However, as of the end of 2024, the US lead had shrunk to 0.3 percentage points. "Around 2015, China put itself on the path to be a top player in AI, and they did it through investments in education," says Selman. "We're seeing that's starting to pay off." The field has also seen a surprising surge in the number and performance of 'open weight' models such as DeepSeek and Facebook's LLaMa. Users can freely view the parameters that these models learn during training and use to make predictions, although other details, such as the training code, might remain secret. Originally, closed systems, in which none of these factors are disclosed, were markedly superior, but the performance gap between top contenders in these categories narrowed to 8% in early 2024, and to just 1.7% in early 2025. "It's certainly good for anyone who can't afford to build a model from scratch, which is a lot of little companies and academics," says Ray Perrault, a computer scientist at SRI, a non-profit research institute in Menlo Park, California, and co-director of the report. OpenAI in San Francisco, California, which developed the chatbot ChatGPT, plans to release an open-weight model in the next few months. After the public launch of ChatGPT in 2022, developers put most of their energy into making systems better by making them bigger. That trend continues, the index reports: the energy used to train a typical leading AI model is currently doubling annually; the amount of computing resources used per model is doubling every five months; and the training data sets are doubling in size every eight months. Yet companies are also releasing very capable small models. The smallest model registering a score higher than 60% on the MMLU in 2022, for example, used 540 billion parameters; by 2024, a model achieved the same score with just 3.8 billion parameters. Smaller models train faster, give faster answers and use less energy than larger ones. "It helps everything," says Perrault. Some smaller models can emulate the behaviour of larger models, says Selman, or take advantage of better algorithms and hardware than those in older systems. The index reports that the average energy efficiency of hardware used by AI systems improves by about 40% annually. As a result of such advances, the cost of scoring just over 60% on the MMLU has plummeted, from about US$20 per million tokens (bits of words produced by language models) in November 2022 to 7 cents per million tokens in October 2024. Despite striking improvements on several common benchmark tests, the index highlights that generative AI still suffers from issues such as implicit bias and a tendency to 'hallucinate', or spit out false information. "They impress me in many ways, but horrify me in others," says Selman. "They surprise me in terms of making very basic errors."

The AI Race Has Gotten Crowded -- and China Is Closing In on the US

New research from Stanford suggests artificial intelligence isn't ruled by just OpenAI and Google, as competition increases across the US, China, and France. The year that ChatGPT went viral, only two US companies -- OpenAI and Google -- could boast truly cutting-edge artificial intelligence. Three years on, AI is no longer a two-horse race, nor is it purely an American one. A new report published today by Stanford University's Institute for Human-Centered AI (HAI) highlights just how crowded the field has become. The institute's 2025 AI index, which collates data and trends on the state of the AI industry, paints a picture of an increasingly competitive, global, and unrestrained race towards artificial general intelligence -- AI that surpasses human abilities. OpenAI and Google are still neck and neck in the race to build bleeding-edge AI, the report shows. But several other companies are closing in. In the US, the fiercest competition comes from Meta's open weight Llama models; Anthropic, a company founded by former OpenAI employees; and Elon Musk's xAI. Most strikingly, according to a widely used benchmark called LMSYS, the latest model from China's DeepSeek, R1, ranks closest to the top-performing models built by the two leading American AI companies. "It creates an exciting space. It's good that these models are not all developed by five guys in Silicon Valley," says Vanessa Parli, director of research at HAI. "Chinese models are catching up as far as performance to the US models," Parli adds, "But across the globe, there are new players emerging in the space." The arrival of DeepSeek-R1 in January sent shock waves through the US tech industry and stock market. The company claimed to have built its model using a fraction of the compute used by US rivals. DeepSeek's debut was also a surprise because the US government has repeatedly sought to limit China's access to the computer chips needed to build the most advanced AI.

AI Report Highlights Smaller, Better, Cheaper Models

A state of the AI industry report shows that 2024 was a breakthrough year for small, sleek models to rival the behemoths The artificial intelligence (AI) race is heating up: the number and quality of high-performing Chinese AI models is rising to challenge the US lead, and the performance edge between top models is shrinking, according to an annual state of the industry report. The report highlights that as AI continues to improve quickly, no one firm is pulling ahead. On the Chatbot Arena Leaderboard, which asks users to vote on the performance of various bots, the top-ranked model scored about 12% higher than the tenth-ranked model in early 2024, but only 5% higher in early 2025 (see 'All together now'). "The frontier is increasingly competitive -- and increasingly crowded," the report says. The Artificial Intelligence Index Report 2025 was released today by the Institute for Human Centered AI at Stanford University in California. If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today. The index shows that notable generative AI models are, on average, still getting bigger, by using more decision-making variables, more computing power and bigger training data sets. But developers are also proving that smaller, sleeker models are capable of great things. Thanks to better algorithms, a modern model can now match the performance that could be achieved by a model 100 times larger two years ago. "2024 was a breakthrough year for smaller AI models," the index says. Bart Selman, a computer scientist at Cornell University in Ithaca, New York, who was not involved in writing the Index report, says it's good to see relatively small, cheap efforts such as China's DeepSeek proving they can be competitive. "I'm predicting we'll see some individual teams with five people, two people, that come up with some new algorithmic ideas that will shake things up," he says. "Which is all good. We don't want the world just to be run by some big companies." The report shows that the vast majority of notable AI models are now developed by industry rather than academia: a reversal of the situation in the early 2000s, when neural nets and generative AI had not yet taken off. Industry produced fewer than 20% of notable AI models before 2006, but 60% of them in 2023 and nearly 90% in 2024, the report says. The United States continues to be the top producer of notable models, releasing 40 in 2024, compared with China's 15 and Europe's 3. But plenty of other regions are joining the race, including the Middle East, Latin America and southeast Asia. And the previous US lead in terms of model quality has disappeared, the report adds. China, which produces the most AI publications and patents, is now developing models that match their US competition in performance. In 2023, the leading Chinese models lagged behind the top US model by nearly 20 percentage points on the Massive Multitask Language Understanding test (MMLU), a common benchmark for large language models. However, as of the end of 2024, the US lead had shrunk to 0.3 percentage points. "Around 2015, China put itself on the path to be a top player in AI, and they did it through investments in education," says Selman. "We're seeing that's starting to pay off." The field has also seen a surprising surge in the number and performance of 'open weight' models such as DeepSeek and Facebook's LLaMa. Users can freely view the parameters that these models learn during training and use to make predictions, although other details, such as the training code, might remain secret. Originally, closed systems, in which none of these factors are disclosed, were markedly superior, but the performance gap between top contenders in these categories narrowed to 8% in early 2024, and to just 1.7% in early 2025. "It's certainly good for anyone who can't afford to build a model from scratch, which is a lot of little companies and academics," says Ray Perrault, a computer scientist at SRI, a non-profit research institute in Menlo Park, California, and co-director of the report. OpenAI in San Francisco, California, which developed the chatbot ChatGPT, plans to release an open-weight model in the next few months. After the public launch of ChatGPT in 2022, developers put most of their energy into making systems better by making them bigger. That trend continues, the index reports: the energy used to train a typical leading AI model is currently doubling annually; the amount of computing resources used per model is doubling every five months; and the training data sets are doubling in size every eight months. Yet companies are also releasing very capable small models. The smallest model registering a score higher than 60% on the MMLU in 2022, for example, used 540 billion parameters; by 2024, a model achieved the same score with just 3.8 billion parameters. Smaller models train faster, give faster answers and use less energy than larger ones. "It helps everything," says Perrault. Some smaller models can emulate the behaviour of larger models, says Selman, or take advantage of better algorithms and hardware than those in older systems. The index reports that the average energy efficiency of hardware used by AI systems improves by about 40% annually. As a result of such advances, the cost of scoring just over 60% on the MMLU has plummeted, from about US$20 per million tokens (bits of words produced by language models) in November 2022 to 7 cents per million tokens in October 2024. Despite striking improvements on several common benchmark tests, the index highlights that generative AI still suffers from issues such as implicit bias and a tendency to 'hallucinate', or spit out false information. "They impress me in many ways, but horrify me in others," says Selman. "They surprise me in terms of making very basic errors."

The AI model race has suddenly gotten a lot closer, say Stanford scholars

The competition to create the world's top artificial intelligence models has become something of a scrimmage, a pile of worthy contenders all on top of one another, with less and less of a clear victory by anyone. According to scholars at Stanford University's Institute for Human-Centered Artificial Intelligence, the number of contenders in "frontier" or "foundation" models has expanded substantially in recent years, but the difference between the best and the weakest has also narrowed substantially. In 2024, "the Elo score difference between the top and 10th-ranked model on the Chatbot Arena Leaderboard was 11.9%. By early 2025, this gap had narrowed to just 5.4%," write Rishi Bommasani and team in "The AI Index 2025 Annual Report" "The AI landscape is becoming increasingly competitive, with high-quality models now available from a growing number of developers," they write. The gap between OpenAI and Google has narrowed even more, with the GPT family and Gemini having a performance difference of just 0.7%, down from 4.9% in 2023. A concurrent trend, according to Bommasani, is the rise of "open-weight" AI models, such as Meta Platforms's Llama, which can, in some cases, equal the top "closed" models, such as GPT. Open-weight models are those where the trained weights of the neural nets, the heart of their ability to transform input into output, are made available for download. They can be used to inspect and replicate the AI model without having access to the actual source code instructions of the model. Closed models do not provide public access to weights, and so the models remain something of a black box, as is the case with GPT and Gemini. "In early January 2024, the leading closed-weight model outperformed the top open-weight model by 8.0%. By February 2025, this gap had narrowed to 1.7%," write Bommasani and team. Also: Gemini Pro 2.5 is a stunningly capable coding assistant - and a big threat to ChatGPT Since 2023, when "closed-weight models consistently outperformed open-weight counterparts on nearly every major benchmark," they relate, the gap between closed and open has narrowed from 15.9 points to "just 0.1 percentage point" at the end of 2024, largely a result of Meta's 3.1 version of Llama. Another thread taking place alongside open-weight models are the surprising achievements of smaller large language models. AI models are typically classified based on the number of weights they use, with the biggest at the mo ment publicly disclosed, Meta's Llama 4, using two trillion weights. "2024 was a breakthrough year for smaller AI models," write Bommasani and team. "Nearly every major AI developer released compact, high-performing models, including GPT-4o mini, o1-mini, Gemini 2.0 Flash, Llama 3.1 8B, and Mistral Small 3.5." Bommasani and team don't make any predictions about what happens next in the crowded field, but they do see a very pressing concern for the benchmark tests used to evaluate large language models. Those tests are becoming saturated -- even some of the most demanding, such as the HumanEval benchmark created in 2021 by OpenAI to test models' coding skills. That affirms a feeling seen throughout the industry these days: It's becoming harder to accurately and rigorously compare new AI models. Also: With AI models clobbering every benchmark, it's time for human evaluation In response, note the authors, the field has developed new ways to construct benchmark tests, such as Humanity's Last Exam, which has human-curated questions formulated by subject-matter experts; and Arena-Hard-Auto, a test created by the non-profit Large Model Systems Corp., using crowd-sourced prompts that are automatically curated for difficulty. The authors note that one of the more challenging tests is the ARC-AGI test for finding visual patterns. It's still a hard test, even though OpenAI's o3 mini did well on it in December. The hardness of the benchmark is affecting AI models for the better, they write. "This year's improvements [by o3 mini] suggest a shift in focus toward more meaningful advancements in generalization and search capabilities" among AI models, they write. The authors note that creating benchmarks is not simple. For one, there is the model of "contamination," where neural networks are trained on data that ends up being used as test questions, like a student who has access to the answers ahead of an exam. Also: 'Humanity's Last Exam' benchmark is stumping top AI models - can you do any better? And many benchmarks are just badly constructed, they write. "Despite widespread use, benchmarks like MMLU demonstrated poor adherence to quality standards, while others, such as GPQA, performed significantly better," according to a broad research study at Stanford called BetterBench. Bommasani and team conclude that standardizing across benchmarks is essential going forward. "These findings underscore the need for standardized benchmarking to ensure reliable AI evaluation and to prevent misleading conclusions about model performance," they write. "Benchmarks have the potential to shape policy decisions and influence procurement decisions within organizations, highlighting the importance of consistency and rigor in evaluation."

Which country is leading the AI race?

A new report has found that the AI race is "tighter than ever" with the United States and China ceding ground. As the global race to become the leader in generative artificial intelligence (GenAI) gains pace, new data shows that it is no longer just two horses in the running. The United States is still in the lead but China is closing the performance gap and Europe is also gaining ground, according to a new report from Stanford University. US-based institutions produced 40 notable AI models, while China produced 15 and Europe produced three in 2024. Despite the number produced, the 2025 Stanford Artificial Intelligence Index weighed up several benchmarks and found that Chinese models reached near parity with the US on two performance markers: Massive Multitask Language Understanding (MMLU), which tests AI's knowledge and problem-solving ability, and HumanEval, which evaluates code generation capabilities. "The race is tighter than ever, and no one has a clear lead," the report authors from Stanford University said. The results come as global leaders say winning the AI race is critical to national security and for advancements in health, business and technology. Meanwhile, companies such as OpenAI, Google and DeepSeek, among many others, are battling it out to build the best AI platforms. Catch up When OpenAI's ChatGPT went viral in late 2022, only it and Google had pioneering AI tech. But fast forward to today and other US companies, such as Meta, Elon Musk's xAI and Anthropic are catching up, the report showed. Another benchmark also showed that China's DeepSeek R1 model ranked closest to OpenAI and Google's models. DeepSeek sparked a frenzy in January when it came onto the scene with R1, an artificial intelligence (AI) model and chatbot that the company claimed was cheaper and performed just as well as OpenAI's rival ChatGPT model. As for the most notable machine learning models in 2024, the top contributors were OpenAI (seven models), Google (six) and China's Alibaba (four), the report showed. France's Mistral AI came in eighth place with three models. Most patents Another marker that the AI race is narrowing is the number of AI publications and patents, which shows China way out in front of the US. The report found that as of 2023, China leads in total AI patents, accounting for almost 70 per cent of all grants. South Korea came in second place and Luxembourg in third, which also stood out as the top AI patent producer on a per capita basis. AI model development is increasingly global, with notable launches from regions such as the Middle East, Latin America, and Southeast Asia, the report added.

Global AI race intensifies as China closes gap in quality; OpenAI, Google hold strong: Report

China has been advancing its AI capabilities with the launch of Alibaba's Qwen Series, DeepSeek's R1, ManusAI and Tencent's Hunyuan Turbo S, among others. This is thanks to China's aggressive investments in AI infrastructure, advanced computing capabilities, and state-sponsored research initiatives.The United States continues to dominate the global landscape of artificial intelligence (AI) model development, producing 40 notable models in 2024 alone. However, China is swiftly narrowing the performance gap, signalling a transformative shift in the global AI race, according to the latest Artificial Intelligence Index Report. China adds quality, US builds numbers While the US has been a dominant player in building top-tier AI models, China has made significant strides in quality. In 2023, there was a double-digit performance gap between Chinese and American models across industry-standard benchmarks, such as Massive Multitask Language Understanding (MMLU) and HumanEval (coding performance). By 2024, this difference had narrowed to near parity. China has been advancing its AI capabilities with the launch of Alibaba's Qwen Series, DeepSeek's R1, ManusAI and Tencent's Hunyuan Turbo S, among others. This improvement is attributed to China's aggressive investments in AI infrastructure, advanced computing capabilities, and state-sponsored research initiatives. The low-cost model, developed within two months and with an investment of less than $6 million, starkly contrasts the $100 million that OpenAI reportedly spent on training its GPT-4 model, ET reported in January. Source: Artificial Intelligence Index Report Top AI organisations: OpenAI, Google, Alibaba The global AI race towards building agentic capabilities and infrastructure has caught the attention of the biggest tech giants and academic institutions. In 2024, OpenAI emerged as the top organisational contributor, releasing seven notable AI models and emerging as a key player in the general-purpose AI systems. Close behind, Google followed with six significant model launches, reinforcing its long-standing leadership in machine learning (ML) innovation. Over the past decade, Google has maintained a dominant position, contributing a staggering 186 notable models since 2014 -- more than double the next player on the list. Source: Artificial Intelligence Index Report Meta and Microsoft have also been prolific, with 82 and 39 models developed over the same period, respectively. Meta on Saturday unveiled its latest suite of open-weight AI models under the Llama 4 family, including two new variants -- Llama 4 Maverick and Llama 4 Scout -- aimed at delivering personalised, multimodal systems. Notably, Alibaba, representing China's growing presence in foundational AI development, ranked third in 2024 with four notable models. This signals a shift in the global innovation landscape, where Chinese firms are not only scaling deployment but also contributing to frontier-level research and model design. Among academic institutions, Carnegie Mellon University, Stanford University, and Tsinghua University have been the most prolific since 2014, with 25, 25, and 22 notable models, respectively. Amplifying AI research & patents In addition to model quality, China leads the world in AI research volume. In 2023, Chinese researchers accounted for 23.2% of all AI-related publications, compared to 15.2% from Europe and just 9.2% from India. China's share has grown steadily since 2016, as European contributions declined and US publication output plateaued. Despite America's ban on the supply of AI chips, China has emerged as the second largest nation in terms of producing AI models across text, images, video and audio. Out of a total of 1,328 AI large language models (LLMs) globally, 36% originated in China, ranking second after the US. While China leads in volume, the US still maintains an edge in influence. American institutions contributed the majority of the top 100 most-cited AI papers over the last three years. AI model development The report highlighted notable achievements from regions such as the Middle East, Latin America, and Southeast Asia -- signalling the rise of a more globally distributed AI innovation ecosystem. France was the leading European nation in 2024 with three notable models. Overall, however, all major regions -- including the US, China, and the EU -- saw a decline in the number of notable models released compared to 2023.