00:00Since the launch of ChatGPT in 2022, America has been commonly perceived as the global leader in AI.
00:07Its industry boasts advanced models, chip production, and large sums of private investor cash.
00:12But China has grown to be a formidable rival with a few advantages of its own,
00:16which include a large pool of engineering talent, lower costs, and a state-led development model.
00:26Last year, China shocked the world with a debut of R1, a model from domestic research firm DeepSeek
00:32that can compete with ChatGPT while only using a fraction of the compute.
00:36In 2025, the value of China's AI industry was estimated at more than $160 billion,
00:43and the country is already home to more than 5,300 AI enterprises.
00:47That includes Baidu, one of China's foremost full-stack AI companies.
00:52In November, I attended the Baidu Annual Conference,
00:55where I spoke to CEO Robin Li about how China's approach to AI differs from the American view.
01:01We do have very different views.
01:04I think the mainstream U.S. people do take it like a Manhattan Project-like thing.
01:13They invest hugely to achieve the so-called AGI.
01:17So that the U.S. will be ahead of China and all the other countries.
01:24We care a lot more about applications.
01:27You know, China is very strong in manufacturing.
01:30We need to manufacture at a low cost, at a very high efficiency.
01:36And we need to use AI to solve those problems.
01:42That's what we care more about.
01:46I'm not even convinced that there is the so-called AGI that one model fits all.
01:52I think right now it's like a pyramid that, you know, most value is realized at the chip level.
01:58Then the model level realizes probably the 110 and the application layer realized even less.
02:05This is certainly unhealthy.
02:09I firmly believe that you have to create much more value at the application layer in order to sustain the
02:20investment in models, in chips and so on.
02:23The access to chips is a critical component of the AI arms race, with the U.S. maintaining a strong
02:29lead over China in chip production.
02:31The Trump administration recently brokered a carve out for NVIDIA to export its second most powerful chips to Chinese firms.
02:38But Beijing is blocking their imports, presumably to boost demand for homegrown equivalents.
02:43In November, Baindu unveiled its latest M100 AI chips and plans to build a supernode capable of supporting millions of
02:51chips by 2030.
02:53We are probably a few years behind on chips, but we're not that far behind on the model level.
02:59And on top of the models, we have lots of lots of application scenarios you cannot find elsewhere.
03:06You cannot find it in the U.S.
03:09So U.S. people don't know, don't even know that they need to solve this kind of problems.
03:14We know, we sometimes have to solve those problems, right?
03:18That's where the value gets created.
03:20So I'm not so worried about the restrictions on the chip, although I very much like to get access to
03:27the most advanced NVIDIA chips.
03:29The Chinese state is deeply involved in the management of its AI industry.
03:33In August, Beijing unveiled its AI Plus initiative, which aims to integrate AI into 90% of the country's economy
03:40by 2030.
03:41In China, new commercial technology isn't permitted until it's been regulated by the government,
03:46a system that advocates like Li says leads to smoother adoption.
03:50Overall, the Chinese government is pro-innovation.
03:55They always say we support your innovation efforts.
04:00But on the other hand, they need to also care about all kinds of concerns from the stakeholders.
04:09And if no regulation says that you can have a self-driving car on the road, then it means it's
04:18not.
04:18This is a little bit different from the U.S.
04:21In the U.S., if there's no regulation that says you cannot have a self-driving car on the road,
04:27then you are allowed.
04:28Especially in states like Texas or Georgia, there's no regulation at all on the robot taxi operation.
04:36But for China, you have to, in a lot of cases, get permission from the regulators.
04:45AGI and foundation model, it's similar.
04:48In China, we don't talk a lot about regulating this kind of things.
04:54But there are actually regulations that are guiding the development of this kind of new technologies.
04:59I asked Li what he would say to people worried about AI taking jobs and displacing humans in the workplace.
05:05Yeah, on the longer term, I think there's a consensus that new technology will create a lot more
05:15job opportunities for people.
05:18But in the near term, we do face the challenge that because productivity improvement caused by AI,
05:28there will be downward pressure for employment.
05:32And we need to find ways to handle this.
05:38In the U.S., people talk about the UBI.
05:41For Baidu, we actually helped many cities to establish those data labeling centers that hire hundreds of people.
05:54And I think going forward, we will be able to create a lot more jobs we have never thought of.
06:03For more information, visit our website.
06:06Well, you may be able to create a lot of data.
06:09And you will see all the data across the country and the latest data.
Comments