Is the market underestimating China’s AI potential?

Talking Heads with David Choa

Daniel Morris: Hello and welcome to the BNP Paribas Asset Management Talking Heads podcast. Every week, Talking Heads will bring you in-depth insights and analysis on the topics that really matter to investors. In this episode, we’ll be discussing the Chinese equity market. I’m Daniel Morris, Chief Market Strategist, and I’m joined today by David Choa, Head of Greater Chinese Equities. Welcome, David, and thanks for joining me.

David Choa: Thank you, Daniel. Thanks for having me here today.

DM: China was one of the top performing markets in 2024 and 2025. Unfortunately, it’s near the bottom of the list in the MSCI EM (emerging markets) index, this year. Tech hardware and semiconductors have done well more or less everywhere, but China’s lagged even there. Non-tech hardware stocks [have] not done so well. If you look at non-tech stocks, in China they’re down year to date, where you’re up about 7% in the rest of the AM. If we think about the lag of Chinese equities versus other emerging markets year to date, is it a trend that could continue through the rest of [20]26 or even into next year?

DC: I think [it] really depends on the momentum of this AI (artificial intelligence) trade, it really relate[s] to the AI momentum. In some of these North Asian market[s] like Korea and Taiwan, they are much more heavily stocked with the technology AI semiconductor stock[s]. At this juncture, AI trade is hot. But China, is accelerating its innovation ecosystem. It has a realistic opportunity to catching up with the US, providing an alternative AI infrastructure supply chain.

A lot of these opportunities are being underestimated by the market. And even [though] the Chinese index has been going nowhere this year, within the market, the innovation or tech focus segment has risen quite a lot – the Star 50 index has gone up by 30%. It’s an indicator that there are a lot of interesting ideas within the market.

DM: One of the concerns a lot of people have with AI is ‘[is] it taking over the world?’ To some degree, you could say it’s already done that in terms of equity markets. We appreciate the dominance of tech stocks, say in the US indices. You highlighted already that, in some emerging market countries, the weight of tech is quite dominant in the indices. When we think about China, how do you manage that dominance of tech, of AI, in terms of asset allocation?

DC: The view of our team is that, despite the US dominance of this AI race, the Chinese portion definitely has a role in terms of our global asset allocators’ portfolio, simply because of the diversification.

As I mentioned, this AI cycle, in contrast to the previous major cycle in the mainframe[s], PC[s] and mobile [phones], is different in that it’s much more extensive. It has a much longer and widespread impact, on the supply side but also on the demand side. And, at this juncture, [the] US is definitely dominating, but China is improving and catching on quickly.

In the future, there will be probably more than one ecosystem in the world, so not only China has a lot of opportunities rising, but its different system could be a hedge against what we already own in the US. From global as[set] allocator point of view, I think diversification and hedging are good reason[s] to start it off, and then within the Chinese opportunity set. China is somewhat of a combination of the US, Korea and Taiwan.

[The] US ha[s] the frontier model. China ha[s] DeepSeek, Kimi, Qingwen from Alibaba, etc. Taiwan ha[s] the semiconductor supply chain, and this is an area that China also ha[s] with Hua Hong, SMIC, and other semiconductor equipment names.

On the memory side, one of the biggest Chinese memory player[s] just listed on the Chinese onshore market. So, China has some of the critical building blocks already in place in the system, and will continue to build out this supply chain along around this critical building block.

Secondly, what is interesting in terms of China’s opportunities in AI is actually while the world has been fixated whether the US capex cycle on AI will be ending or not ending, China has been at least one to two year[s] late to the party. Plus, they have been under-investing.

For example, all the Chinese hyperscaler[s] combined, their AI cap rate is less than 1/3 of what the US have been investing cumulatively. So, they have to catch up. That’s the easy part.

The second part is domestic substitution. Because of geopolitical tension[s], no matter what the US cycle is, China is quite determined to replace all this US leading technology with the domestic system. So, whether the US tech cycle has ended or not, Chinese domestic substitution will be a telling effect for the next decade.

The last point is disruption. A few weeks ago, there [wa]s a shock to the US market because of the rise of Kimi (AI chatbot and large language model series) being viewed as almost on par with the performance of the US frontier model, leading to the question whether such high capex intensity in [the] US can continue. If there’s a disruption, China could very well be the source.

For all these reason[s], we see more and more opportunities not only in tech but also in industrial and biotech. So, the opportunities are really broadening, but more importantly, they could be actually a source of the disruption to the US and are definitely a good source for diversification in a global allocation.

DM: You mentioned DeepSeek, the big surprise a year ago. You also mentioned Moonshot’s Kim[i] EK3 model. They’re examples of ‘you really don’t know when or where the next shock is going to come from’. More broadly, we appreciate the potential for disruption coming from AI. What are the areas you see in China that are doing well during the AI era, and where potentially China is being underestimated by the market?

DC: I think the biggest question mark is ‘can China do it’? China is lagging behind [the] US in some aspect[s]. However, China has other advantages that people overlook. One is power. China has diversified power. Already, 1/3 of power generation and 60% of installed capacity are from renewable energy.

Another source of the edge for China to catch up in the AI race is the talent pool. Every year there are five million STEM (science, technology, engineering and mathematics) graduate[s], and there’s a lot of leading Chinese scientists and engineers scattered around the world, many of them in the leading position of key US tech companies.

The third advantage the market may overlook is China not only has a huge consumption market, it also has one of the largest industrial base[s] in the world. What it means is [with] industrial manufacturing and consumption they can create a firewall effect, feeding data to one another that could accelerate the catch-up of the Chinese in the AI infrastructure build and application.

What is also interesting about the market opportunities in China – we have to go back to the the mobile internet era. In the early days of 4G, China does not have a very good homegrown mobile layer. But fast forward to 5G, they are catching up, and now becom[ing] a leading force. But what is also interesting back then is actually the application once the infrastructure layer is okay, even it’s not perfect.

Application in mobile internet, like e-commerce – they catch up very quickly. They have a bit of everything now in place. So, [the West] can be assured that they will catch up very quickly, not only on the infrastructure layer, but more importantly they intend to run fast on the application layer – not only limit[ed] to the digital world, just Chatbot or enterprise software, but extend[ing] into the physical world – autonomous driving, humanoid, robotic, etc.

These are the things that China will want very quickly. China may have a limited source of success in zero to one breakthrough, but has been always good from one to 10 – the scaling –combining a little bit of everything, and then to make the existing breakthrough from the others even better. China will run fast in the infrastructure layer but may even run faster in the application layer, which means the opportunities in China will extend beyond basic AI infrastructure. We are sure we will see many of these opportunities come to market in the short future.

DM: Certainly, anyone who travels to China leaves amazed and impressed by the degree of innovation, and that’s probably a good segue to summarise some of your key points.

China’s innovation ecosystem is accelerating and, you believe, underestimated by the market. The opportunity set within the innovation segments are broadening, and that opportunity depends less on [the] US AI capex cycle and more on China’s own technological breakthroughs and import substitution.

And finally, you highlighted that China has its own advantages and edges in power infrastructure, talent pool, and a large mix of both industrial and consumption base as a source of a positive feedback loop, which will allow the country to close some of these gaps more quickly. Well, David, thank you very much for joining me.

DC: Thank you, Daniel. It’s my pleasure.

DM: That’s it for this week’s episode of Talking Heads. If you would like more information about our capabilities in Chinese equities, please reach out to your asset management contact or check out Viewpoint, our website for investment insights at viewpoint.bnparibas-am.com.

We recommend subscribing to Talking Heads on your favourite podcast channel, such as YouTube or Spotify. You’ll receive your podcast episodes every week. If you like Talking Heads, leave us a positive review and a nice rating.

You’ve been listening to the BNP Paribas Asset Management Talking Heads podcast with me, Daniel Morris, and David Choa, Head of Greater Chinese Equities. Please do join me next week. Until then, take care.