In partnership with
Stop making AI decisions in the dark.
Leadership is asking: are we getting value from AI? Which tools are worth the spend? Where are we exposed? Right now, most teams have no idea.
Harmonic Security Usage Explorer changes that.
You get a complete picture of how your organization uses AI, automatically categorized into custom tasks and use cases.
You’ll see the projects being worked on, who’s using what tools, where AI investments are driving value, and where employees are engaging in risky behavior.
CIOs can rationalize spending and cut wasted licenses. CISOs can pinpoint where risk exists and neutralize it. AI committees can show exactly how their efforts are paying off.
By March 2026, leading U.S. and Chinese AI models were separated by only 2.7 percent. The two countries had traded positions several times since early 2025. That makes a simple question about which country has better AI harder to answer.
The Model Performance Gap Has Narrowed
The United States held a wide model performance advantage during 2023. That difference fell sharply through 2024 and 2025.
Stanford's 2026 AI Index found that Chinese and American models now operate within a narrow performance range. In March 2026, the leading U.S. model scored 1,503 on Arena. The leading Chinese model scored 1,464.
This difference matters, but it no longer defines the competition. Cost, reliability, reasoning quality, context handling, and specific tasks increasingly affect model choice.
The United States Still Produces More Notable Models
American institutions produced 59 notable AI models during 2025. Chinese institutions produced 35.
The U.S. also retained an advantage in patents judged to have higher impact. Its position remains tied to large technology companies, cloud infrastructure, advanced chips, and private capital.
The infrastructure difference remains large. Stanford counted 5,427 data centers in the United States, more than ten times any other country.
AI Insights That Turn Your Data Into Bigger Profits
Your store is full of hidden growth opportunities. StoreClaw analyzes your Shopify and Amazon data, identifies the highest-impact actions, and helps you grow revenue while improving margins. Start free today with 10,000 bonus tokens. No credit card required.
China Leads Several Research Measures
Model counts tell only part of the story.
China leads in AI publication volume, citations, and patent grants. Its share of the world's 100 most cited AI papers increased from 33 in 2021 to 41 in 2024.
China also accounted for 54 percent of global industrial robot installations in 2024. This shows how its AI development connects with manufacturing and physical automation.
Investment Remains Very Different
Private AI investment in the United States reached $285.9 billion during 2025. China's recorded private investment was $12.4 billion.
Stanford warns that this comparison understates Chinese spending. Government guidance funds play a larger role inside China's technology financing system.
The numbers still show different development structures. American AI relies heavily on private companies and capital markets. Chinese development combines private firms, state funding, research institutions, and industrial policy.
Open Models Are Another Point of Difference
Chinese labs have become increasingly visible in open weight AI. DeepSeek, Alibaba, and other Chinese developers have released models that compete close to American proprietary systems.
Stanford researchers wrote in August 2026 that recent Chinese open weight models had moved closer to leading American models. Open weights also allow developers to run, study, modify, and deploy models outside the original provider's hosted service.
There Is No Single AI Leader
The United States remains ahead in private investment, infrastructure, and notable model production. China holds stronger positions in research volume, patent counts, industrial robots, and parts of the open weight ecosystem.
Model capability itself is becoming a weaker dividing line.
The more useful question is becoming specific. Which system performs better for the task, language, cost, deployment method, and reliability requirements involved?
This newsletter cannot settle every part of that decision. It works better as a regular research check on what has changed. The useful work happens afterward, through testing models, reading evidence, and watching results over time.
Free email without sacrificing your privacy
Gmail tracks you. Proton doesn’t. Get private email that puts your data — and your privacy — first.
