China’s Secret Lead: Are They Winning the AI Race?
Chinese AI models are gaining ground fast, powering top apps like Pinterest and Airbnb, proving open-source innovation may outpace US tech.
Every month, millions of users flock to Pinterest in search of the latest styles.
A page called “The Weirdest Things” is full of quirky ideas to inspire creative people. Crocs used as flower pots. Cheeseburger-shaped eyeshadow. A gingerbread house made of vegetables.
But what potential shoppers might not know is that the technology behind it isn't necessarily made in the US. Pinterest is experimenting with Chinese AI models to improve its recommendation engine.
“We’ve essentially turned Pinterest into an AI-powered shopping assistant,” company boss Bill Ready told me.
Of course, the San Francisco-based company could have used any number of American AI labs to power its behind-the-scenes operations.
But since the launch of China’s DeepSeek R-1 model in January 2025, Chinese AI technology has become a significant part of Pinterest.
Ready considers this so-called “DeepSeek moment” a major breakthrough.
“They decided to open-source it, and that kicked off a wave of open-source models,” he said.
Chinese competitors include Alibaba’s Qwen and Moonshot’s Kimi, while TikTok owner ByteDance is also working on similar technology.
Pinterest’s Chief Technology Officer Matt Madrigal said the advantage of these models is that companies like his can download and customize them for free – unlike most models offered by American competitors like OpenAI, which makes ChatGPT.
“The open-source technologies we use to train our in-house models are 30% more accurate than leading off-the-shelf models,” Madrigal said.
And these better recommendations come at a much lower cost, he added, sometimes ninety percent less than using the proprietary models favored by American AI developers.
'Fast and Cheap'
Pinterest is likely not the only American company relying on Chinese AI technology.
These models are gaining popularity across a wide range of Fortune 500 companies.
Airbnb CEO Brian Chesky told Bloomberg in October that his company was "heavily" reliant on Alibaba's Qwen to power its AI customer service agent.
Software Services
 |
Start your own gaming platform for just $3,000 |
He cited three simple reasons – it's "very good," "fast," and "cheap."
And the evidence can be found on Hugging Face, the platform where people go to download pre-trained AI models – including those from major developers like Meta and Alibaba. Jeff Boudier, who builds products on the platform, said it's the cost factor that's driving younger startups to look at Chinese models instead of American ones.
"If you look at the top trending models on Hugging Face – the ones that the community downloads and likes the most – typically, Chinese models from Chinese labs occupy many of the top 10 spots," he told me.
"There are some weeks where four out of the top five trending models on Hugging Face are from Chinese labs."
In September, Qwen overtook Meta's Llama to become the most downloaded family of large language models on the Hugging Face platform.
Meta released its open-source Llama AI models in 2023. Until the release of models from DeepSeek and Alibaba, they were considered the best option for developers working on custom applications.
Read Also
But the release of Llama 4 last year disappointed developers, and Meta is reportedly using open-source models from Alibaba, Google, and OpenAI to train a new set of models expected to be released this spring.
Airbnb also uses multiple models, including US-based ones, and securely hosts them within the company's own infrastructure. According to the company, data is never shared with the developers of the AI models they use.
Chinese Success
In 2025, the prevailing view was that despite billions of dollars spent by American tech firms, Chinese companies were threatening to overtake them.
“That’s not the story anymore,” Baudrier said. “Now, the best model is an open-source model.”
A report published last month by Stanford University found that Chinese AI models have “caught up with or surpassed” their global counterparts – both in terms of what they are capable of doing, and how many people are using them.
Former UK Deputy Prime Minister Sir Nick Clegg said he felt American companies were focusing too much on the pursuit of AI that could one day surpass human intelligence.
Last year, Sir Nick resigned from his position as head of global affairs at Meta, the developer of Llama. His boss, Mark Zuckerberg, has pledged to spend billions of dollars on achieving what he calls “superintelligence.”
Read Also
Some experts are now calling these ambitions vague and poorly defined – giving China the opportunity to dominate the open-source AI space. “That’s the irony,” Sir Nick said. In the battle between “the world’s largest dictatorship” and “the world’s largest democracy” – China and the US – China is “doing more to democratize the technology they are competing for.”
On the other side of the world, American companies like OpenAI are under immense pressure to increase revenue and become profitable – and they are now turning to advertising to achieve this.
The company launched two open-source models last summer – its first in years. But it has focused most of its resources on its proprietary models to generate revenue.
OpenAI CEO Sam Altman told me in October that the company is rapidly investing in securing more computing power and infrastructure deals with partners.
He said, "Revenue will grow very quickly, but you should expect us to make massive investments in training, the next model, and the model after that."
Thank you for reading this content.
What's Your Reaction?
Like
0
Dislike
0
Love
0
Funny
0
Angry
0
Sad
0
Wow
0
