There are several different reasons why claims that “China has won the AI race” continue to circulate. Some are deliberate propaganda, while others arise from genuine uncertainty, commercial incentives, or the tendency to oversimplify a very complicated competition.
Here are the main mechanisms.
1. Strategic messaging by the Chinese government
The Chinese Communist Party has strong incentives to project technological superiority.
Messages such as:
- “China is overtaking America.”
- “US sanctions failed.”
- “Chinese AI is now the world’s best.”
serve multiple purposes:
- reassure domestic audiences,
- attract investment,
- discourage rivals,
- convince developing countries that China is becoming the inevitable technological leader.
This is common behavior for major powers. During the Cold War, both the United States and the Soviet Union regularly exaggerated their own strengths.
2. Western media often overreacts to milestones
Suppose a Chinese company releases a strong model.
Headlines may become:
“China has caught up.”
or even
“The AI race is over.”
But AI is not a single race.
It consists of many races:
- foundation models
- chips
- cloud infrastructure
- robotics
- scientific AI
- software ecosystems
- developer tools
- enterprise deployment
- military applications
- talent
- capital
- energy supply
Leading in one category does not imply leading overall.
3. People confuse one impressive model with an entire ecosystem
Imagine two countries.
Country A builds:
- five excellent models,
- the world’s leading AI chips,
- most developer tools,
- most AI startups,
- most cloud infrastructure.
Country B releases:
- one excellent model.
Many people conclude:
Country B has won.
This is like saying one Olympic gold medal means winning the entire Olympics.
4. Cost efficiency gets mistaken for overall superiority
Sometimes Chinese labs demonstrate:
- cheaper training,
- lower inference costs,
- clever engineering.
Those achievements are real and important.
But cost efficiency is only one dimension.
Other dimensions include:
- frontier research,
- original algorithms,
- semiconductor technology,
- software ecosystems,
- global adoption,
- scientific impact.
5. Information operations amplify favorable narratives
Various actors—including state media, affiliated influencers, and online accounts—may amplify stories that portray China as technologically unstoppable. Researchers have documented coordinated influence campaigns around geopolitical and technology topics, although the scale and effectiveness vary over time.
The goal is not necessarily to convince everyone.
Often it is simply to create uncertainty:
“Maybe China is already ahead.”
That uncertainty alone can shape public perception.
6. The AI race has no finish line
Unlike the Apollo 11 Moon Landing, AI has no single moment that determines a winner.
Today’s leader may not be tomorrow’s.
The frontier changes every few months.
Where does the United States still have major advantages?
As of 2026, the United States retains substantial strengths in several areas:
- frontier foundation model research
- AI chip design
- hyperscale cloud computing
- venture capital
- university research
- global software ecosystems
- concentration of leading AI companies
- access to cutting-edge semiconductor manufacturing through international partnerships
China, meanwhile, has notable strengths in:
- manufacturing scale
- robotics deployment
- abundant engineering talent
- rapid commercialization
- large domestic datasets in some sectors
- government coordination
- AI adoption across industry
These are meaningful advantages, but they are different from “winning” outright.
Why the claim is appealing
Humans naturally prefer simple stories.
Instead of saying:
“Competition across compute, algorithms, energy, talent, manufacturing, and institutions remains dynamic and mixed.”
people prefer:
“China won.”
or
“America won.”
Reality is much less tidy.
One observation connects to themes we’ve discussed before. In recent conversations, you’ve emphasized the distinction between 碎片信息 (fragmentary information) and the deeper patterns beneath it. “China has won the AI race” is a classic example of a fragment that can become a compelling narrative. A more informative question is not who is winning today, but which capacities determine long-term leadership. On that question, the evidence suggests an ongoing competition with different strengths on each side rather than a settled outcome.
