Differentiated Understanding

Joining me today is Chim Lee, Senior Analyst at the Economist Intelligence Unit. He works in EIU’s China and Asia teams, and is based in the company’s Beijing and Hong Kong offices.He leads EIU’s research on China’s advanced technologies, Climate change, Energy, Semiconductors, and Artificial intelligence, and also covers how China’s industrial policies link up with the broader diplomatic and macroeconomic dynamics.Our conversation starts with China’s newly announced 15th Five-Year Plan proposal, which reveals the country's next priority and how it may impact its economy, society, and trade relations with the rest of the world.We then dove into the current involution 内卷 issue, particularly zooming in on the solar and EV sectors. Then we look at the data center build-out driven by the AI boom and how local and regional governments are making sure involution does not hamper this sector.Finally, Chim reflects on his work and his analysis of China’s economic planning and innovation direction.--In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.They delve into the strategic priorities outlined in the plan, including self-reliance in technology, maintaining manufacturing dominance, and the role of private and public sectors in driving economic growth. The conversation also touches on the challenges of overcapacity and the evolving landscape of China’s international cooperation.--* China’s Five-Year Plan signals strategic priorities.* Focus on self-reliance in technology and innovation.* Maintaining manufacturing dominance is crucial.* Private sector plays a key role in economic growth.* Overcapacity remains a challenge in various sectors.* International cooperation is evolving in China’s strategy.* AI and new energy are critical emerging industries.* China’s economic planning involves both public and private sectors.* The plan addresses geopolitical tensions and trade flows.* China’s approach to technology is both strategic and adaptive. Get full access to AI Proem at aiproem.substack.com/subscribe

In this conversation, I spoke with Tom Nunlist from policy consultancy Trivium, about China’s AI Plus plan and its implications for the economy and society. We discussed the role of digital infrastructure in AI adoption, the transformation of production relations, demographic challenges, and the government’s role in connecting academia and industry. The conversation also covers the complexities of navigating China’s regulatory landscape, municipal and provincial implementations of AI policies, and the measurement of AI’s economic impact. Tom shares insights on how MNCs can better align corporate strategies with government objectives during the AI growth era, and talks about the emerging AI pilot zones and how China balances between innovation and regulation. Tom Nunlist is the Associate Director of Tech and Data Policy at Trivium, a leading China policy research consultancy. Tom’s work explores the intersection of politics and technology, with a focus on data and artificial intelligence. His hands-on consulting work with Fortune 100 clients covers policy analysis, risk assessment, government relations, and communications.In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.Chapters04:27 Understanding China’s AI Plus Plan10:55 Transforming Production and Society with AI15:54 Government’s Role in AI Development24:59 Measuring AI’s Economic Impact27:12 Local Adaptation in Policy Implementation28:01 Understanding Chinese Policymaking for MNCs28:59 Aligning Corporate Goals with Government Objectives31:19 AI Pilot Zones and Innovation Hubs33:26 Promising Use Cases for AI Adoption35:56 Balancing Innovation and Regulation in AI42:52 Shifts in Government Priorities for Technology45:56 Tracking Real AI Diffusion in the Economy48:57 The Skills Gap Created by AIAI Generated TranscriptGrace Shao (00:00)Joining me today is Tom Nunlist, Associate Director of Tech and Data Policy at Trivium, a leading China policy research consultancy. Tom’s work explores intersection of politics and technology with a focus on data and artificial intelligence. His hands-on consulting work with Fortune 100, clients, covers, policy analysis, risk assessment, government relations, and communications. Tom, it’s so great to have you here and it was lovely meeting you online actually a couple weeks ago at one of the panels we were on together.So today, will unpack China’s AI Plus plan, what it means for the real economy and how AI governance is viewed by China, sorry, viewed in China and compare that to what’s happening really in the US. But to start, tell us about Trivium and what’s your own professional journey. How did you kind of end up in Shanghai?Tom Nunlist (00:45)Cool, thanks, Grace. That was a nice introduction and likewise, good to meet you and good to be here on your podcast. Definitely flattering. I’ve seen your upcoming guest list and lots of exciting personalities coming up to be on the podcast. So yeah, a little bit about Trivium. We were founded ⁓ in, I think, 2017. So we’ve been around about eight years now. We are a China-focused, or right now China-only policy consultancy.⁓ And so we really our kind of like value ⁓ is that we really know how the sausage is made here in terms of policy and politics in China and we help our clients mostly multinationals and investor clients understand that. So, you know, for example, a new policy like AI plus, you know, comes out, you know, we can come in very quickly and, you know, help inform our clients, you know, what this is, where it comes from, its overall context.and then forming scenarios for how it’s gonna play out and kind of what they might wanna do. As you mentioned, I’m on our tech team, but we cover a lot more ⁓ than tech, really kind of the whole nine yards of ⁓ policy making, be it from economics to labor to kind of everything in between.Tom Nunlist (02:01)Yeah, as for for myself, I think I have a pretty typical China story, you know, insofar as, you know, long time expats. I came here in 2008, you know, more or less on a lark as a study abroad students, you know, to figure it all out and, know,then life happened, got interested in it. I moved here permanently in 2013. My undergrad background is in journalism. So I studied here for a bit. Then I worked at a business review magazine and then eventually kind of made my way into the consulting space. Not too much of a very strict career plan, but again, one thing sort of leading to the next and here we are.Grace Shao (02:43)Awesome. So I think we’re going to go straight into it. What everyone’s interested right now is in the AI Plus plan. So that rolled out in August this year, believe, late August. It’s quite new. think people are still trying to understand what it really means. So the Chinese State Council published a high level paper that was basically pushing all sectors to really embrace AI.It’s said to be the most comprehensive blueprint for AI development domestically, and even touches on China’s international ambitions or diplomacy as well. So to start off with, can you just tell us, high level, what is this really about? How do we understand this policy?Tom Nunlist (03:24)So this is the second high level AI policy to come out of China. The first was some years ago already back in 2017. That was about, you know, it was about the new generation of AI. AI Plus, the concept has been around for two years already now. It was originally announced at the two sessions. Hopefully our listeners know what that is. an annual meeting like that sets policy. It was announced two years ago, talked about again this year, but you know, not manydetails were revealed about it. There was some assumptions which they were correct that it would be a bit like a former policy from 2015 called Internet Plus, was kind of following in that same vein. And just to sort of like set the stage of like what a document like this is, so it’s called a State Council opinion, you know we’re referring to it as a plan, it’s called Plan in the name, but it’s really a sort of like high levellike political document that is setting the priorities for what the nation wants to do, right? Like here’s the direction we’re gonna move in. Here are some like broad, you know, over the horizon kind of KPIs. This is where we all wanna go. ⁓ And then, you know, the plan will cascade kind of down from there. We’ll get more details over time. In terms of the name, AI Plus, that’s AI Plus or added to everything.So everything in the economy, in all sectors, in society, we wanna see, the government wants to see AI make its way into there to get the most use out of it, to really ultimately transform the way that the economy and society works. It’s big, it’s a big vision is the answer.Grace Shao (05:04)think it’s really interesting you mentioned Internet Plus because I remember when that came out. So you said roughly 10 years ago, It was really embraced by every part of the economy and society. So have you seen any attitude changes or shifts or how people view AI Plus just being on the ground in China right now?Tom Nunlist (05:22)Well, in terms of views on the ground, like in terms of people talking about it publicly, not a whole bunch. I think what’s really happened here is these both internet plus and AI plus are responding to things that are already happening. So the internet certainly did not arrive in China in 2015. was ⁓ like.Grace Shao (05:42)Yeah.Tom Nunlist (05:43)very, very much going strong and was already one of the world’s leading digital economies at that time. And so what it was really seeking to do is kind of take the momentum for things that were already happening and push that further. So obviously, in 2015,you know, the consumer internet, know, Alibaba, Taobao, like those type of, certainly WeChat Pay was introduced in 2013. You know, these were making waves, making big changes in the way that society just kind of basically works. And then Internet Plus was like, yeah, let’s take this momentum and apply it to everything else. Let’s have Internet Plus healthcare. How can we use the Internet there? You how can we use it in government services? And again, AI Plus is sort of doing the same thing. You know, the, you know, this is an introducing the AI wave to China, the AI wave is here, it’s happening, everyone’s using it, everyone’s excited. And so this is getting behind that momentum that is naturally already here and attempt to build a policy framework around it and like, yeah, really, where are we going with all this momentum, right? What are we aiming to achieve?Grace Shao (06:52)Yeah, I think actually on that, I am curious, has China’s highly digitalized society and the infrastructure made AI implementation or diffusion any easier in your eyes? Or how has that kind digital infrastructure played a part in just the mass consumer adoption of AI we’re seeing right now in China?Tom Nunlist (07:05)Mm-hmm.Yeah, it’s a huge part. mean, from just the consumer side, China is like the US or Europe, just an extremely connected society. Everyone, even in the most remote places in the country, has their smartphone, has probably even has like 5G.WeChat is something of a national infrastructure at this point. It’s a messaging app that everyone uses for work and life. It really is absolutely indispensable. And so having that infrastructure already there, or having everybody with a phone in their pocket automatically makes these tools accessible. I think...any age, any person you might come across, you know, do they have DeepSeek on their phone? Chances are, yes, they probably do. On the back end, which I think is, you know, just as ⁓ important. So the past few years, or as we all know now, right, the...know, the biggest part of the biggest spend of the AI boom is building out, you know, massive, massive data centers, right? And making that kind of infrastructure work. It’s a huge race right now in the United States. And so there was already a national plan to have ⁓ a nationwide network of data centers, you know, put in place as kind of before this big AI wave.It’s a bit to do with some broader reasons of internet and energy and having some of this infrastructure in place. ⁓ Actually, in terms of energy, that’s one of the ways that I think a big leg up that China has in terms of the US is the amount of energy infrastructure it has built out compared to other parts of the world. So they’re ready to sort of do this, where I think other places maybe a little bit less so.Grace Shao (08:57)Yeah, and I think we can kind of talk a little bit more about the East data, West compute and all these different government initiatives that’s really boosted the data center built out. like, to your point, priority even the AI boom. But actually, let’s take a step back and kind of look at how the policies really affect a society, right? I think in September, one of your colleagues Kendra, her shape for a road, a blog post saying,The state council’s AI Plus directive is to reshape the paradigm of human production and life. When I read that, was like, what does this mean? It seems kind of crazy. Like, are we going to make AI babies now? What does it mean to promote a revolutionary leap in productivity and profound changes in production relations and accelerate the formation of a new intelligent economy? So how do we kind of like break this down? What does it mean? And, you know, when we were chatting prior to this podcast recording, you said this is a grandiose term.It’s three-shaping human production, but what does that actually mean? Like, is this quite literally reproduction? Like, how do we understand that?Tom Nunlist (09:54)Well, I don’t know, hopefully reproduction will stay traditional. in terms of, know, these types of policies sometimes have these like really big grandiose framing. know, again, back to what I said earlier, the point of this, it’s a political document at the end of the day. It’s establishing a vision, right?and the promise of AI, which is not... ⁓news to a Western audience is that it will be transformative of society that will kind of like change sort of how things work. This specific language that they’re using there, like talking about transforming production, you know, that’s a bit in the sort of like communist Marxist language, you know, of China. And then the context that it’s kind of living in now as well, ⁓ there’s this like really big deep sense of urgency in China of like kind of likethe need to move from ⁓ an economic model that is waning, that sort of reliance on labor intensive ⁓ economy and land sales and things like that into a ⁓ new area where they’re get new types of growth, new and better growth, the switch from quantity or quantity.⁓ to quality. These goals were kind of already there. There’s another, you know, wonky Chinese policy term called new quality factors, new quality production factors, and what are all of these, you know, types of new things, you know, ⁓ AI, self-driving cars, and so on. And...it wants to leverage these into making new growth opportunities happen, basically.Grace Shao (11:33)Yeah, and I think you touched on one thing, which is like, you know, the traditional economy is very like labor heavy and it really relied on just the mass population of the mass workforce, right? But what we’re seeing right now in China, but not only China, a lot of like actually very developed economies across Asia, including Japan and South Korea right now.is that there are just not enough people. Like the population is declining, people are not willing to have children, right? And kind of given that backdrop of an aging population, a shrinking population, what is kind of, I guess, the goal from the government when we look at the labor force? And how will AI and technology play a role in that? Are we really just going to see like robots implemented or is it more automation? Or how do we understand this? Yeah.Tom Nunlist (12:22)think in some cases, we will see robots replacing physical workers. ⁓ But I think that’s the smaller part of the story. The bigger part of the story is this broader question of actually just avoiding the middle income trap. And so in order for China to take care of its aging population, to sort of weather this big demographic shift that is happening now,no matter what, even if birth rates double or triple next year, it’s gonna happen. And the way to do that, that the government sees is to raise people’s incomes per capita.and do that very quickly over the next 10 years, More profitable companies have more prosperous people, have a bigger tax base. And so that the country is just able to deal with this challenge as it emerges. Again, it’s inevitable. And so back to this new quality production factors or this transformative effect that AI is gonna have, the transformative effect is that it will be a productivity multiplier, right?enabled everyone and companies from big to small to be vastly more productive and vastly more valuable and really help China earn enough and become wealthy enough, maybe right before it ages too much. So I think a bit indirect there, but it’s about the whole economic story together.Grace Shao (13:51)Yeah, and I think the whole approach to AI development and progress has been extremely pragmatic and economic driven for China, which is a bit different from what I get the sense in DC and even for sure Silicon Valley. Actually, on the topic of what you just mentioned, which is the government’s role in promoting companies and companies’ profitability, I have heard of this thing where the government is playing a role becoming a networker between the academics.Tom Nunlist (14:00)Yeah.Grace Shao (14:19)and the researchers and the companies. And I think for the audience, a lot of people sitting in West, we know about Alibaba Tense and Huawei, these mega big tech companies having talent schemes, quite similar to how basically there’s campus recruitment for like Meta and Google, whatnot, right? But how is the government now playing a role for SMEs or even smaller companies in terms of how are they connecting talent and...⁓ policy people and kind of the resources in the public space and the private space.Tom Nunlist (14:51)This is a great question, think, and a really important thing that’s emerged just over the last couple years. It’s not just AI, it’s sort of like all areas of science, technology, and engineering. But what it seeks to do is to bridge the corporate world and the academic and research world in a better way, right? So you can have like...⁓ needs and talents and coms flowing both ways. So for example, this might be setting up round tables or some kind of like platform or any kind of mechanism that brings these parties together. So going in one direction from corporates, setting up links with universities so they can go into departments and say, hey, we’re doing biosciences, we’re doing AI, we’re doing you know, some type of metallurgy, you know, these are the types of talents we need. And can you focus on that? Can you help get us, you know, train that talent that we need?or going the other way, having researchers see what’s going on in the corporate world and having a solution for that, or green fielding their own research, right? They’ve been doing this for a state institution or a university, and now they wanna take it out of there and find the right entrepreneurial partners to do that with.Right. You know, as you mentioned, know, like large companies have done this sort of thing for for a very long time and have prospered, you know, because of those links. mean, indeed, I mean, a lot of the American tech giants, you know, came. That’s a famous story. It came out of a university or dropped out of a university, you know, and now, you know, maintain those links. It’s same in China, but, know, that’s a lot harder to do if you’re an SME. I mean, everything’s harder to do if you’re an SME because you don’t have the resources. Right. So providing that meeting place,facilitating that is I think a really important program and one that I’m pretty confident will see solid results in the next couple of years.Grace Shao (16:49)So what agency or what government entity is actually helping facilitate that kind of meeting right now? And this question is to lead to the next question, which I’m going to actually ask now, which is, for me, I’m not a policy person. I’m getting confused when I read these papers, right? Because there’s the NDRC, which is in charge of the economic planning.Then MIIT, which is in charge of the information technology, the ministry, the CAC, which is a cybersecurity regulator, right? Then there’s the MOST, and then there’s a party central science technology commission. There’s just so many of these government agencies that seems to be all involved in pushing the progress of AI technology at this point, as well as being a regulator for safety and policy work, right?Could you kind of just break that down? Who’s in charge of what?Tom Nunlist (17:39)All right, okay, let’s just address that one, because that’s like a pretty big question. So for this, for the AI Plus plan in particular, the main administrative body for this is the NDRC. So for those of you don’t know, that is the macroeconomic planner. They’re in charge of kind of like setting the big direction.of the ⁓ economy, right? So NDRC is in the coordinating role of this plan, right? So from there, it’ll go to the other ministries of the state council, some of whom you’ve just mentioned, right? So the industrial ministry, that’s MIT, science and technology ministry, most. ⁓⁓ CAC, the cyberspace administration, all across the board. And those ministries will be in charge of taking the big idea and making it specialized or setting specific goals for their various sectors. And we’ve already seen that happen, actually. just a couple of weeks ago, the National Energy Administration, the NEA, came out with the very first ministerial AI Plus plan, which is AI Plus Energy.And we’ll skip most of the details there, but suffice it to say, it is gonna use AI to help make the energy transition happen, which is very cool. From there, it will cascade down further into localities. And localities is really, that’s where the rubber meets the road and where all of the action happens. So we’ll see cities, they’re already AI plus plans.There’s one in Beijing and Suzhou, those are explicit. And then like Shanghai has one basically, although it’s not called AI Plus, but they have one as well. Interestingly enough, those also actually predate the national plan, which is something that kind of happens in China at various points. And so a lot of the like actual like funding decisions and a lot of where the funding comes from will be at the local level.And then there’ll be like, you know, a national pool of money as well that will like help support those, right? So, you know, it’s a top, you know, people say, China’s a top-down system. That’s of course true. And what I just described is how that top-down system works, right? So from the central planner down to ministry needs, down to local level, which has all of those ministries at the local level and, kind of being funded ⁓ from there. And then of course there’s like special national projects here and there.Grace Shao (19:59)So I’m just trying to understand this. In the RSC, the economic planner basically makes a big grand plan and they push out the AI Plus that we’ve been talking about that was pushed out in August. But a lot of the execution that’s done is actually trickled down into localities like the local governments, the provincial governments, city governments, whatnot. And so something like, just taking this as an example, something like the facilitation of maybe a researcher at Tsinghua meeting private company for potential, let’s say commercialization plan, that could be actually led by say the Beijing Education Department or how does that, I just wanna understand how to execute that works. Okay.Tom Nunlist (20:36)Yeah, yes, yes, yes.Those might exist at different levels, I’m not sure. But yeah, the local level would certainly be implementing stuff like that. Or in another more sort of ⁓ direct way, Shanghai has money now where it can like say, companies that are in AI space are eligible for X amount of money.funding for their first year, right? And like that funding decision, that’s made at the local level at Shanghai.Grace Shao (21:03)And that will be decided, I guess, by what the city might mean. So each city, each province, given their strong, they have their own economic factors, right? Like for example, like I think I was researching Harbin, like, you know, people think it’s just like a really cold place for the ice festival, but actually it’s an industrial city with a lot of legacy in robotics, traditional robotics, mechanics, industrial machinery. So their money might be put into developing physical AI.Tom Nunlist (21:12)Yes.Grace Shao (21:29)like embodied AI, right? And then maybe in Shanghai, we’re thinking about like maybe consumer driven products, right? Like just, just kind of high level thinking, but that that that’s kind of what happens. ⁓ So I want to understand how does the KPI work then, like in terms of like, how do we understand, I guess, how these, because what I’ve heard also is like these cities to cities, compete with each other, they compete for talent, they compete for, like, bringing in different businesses, how does that work? And then in terms of like, how do they actuallyGrace Shao (21:59)a measure, right? Like the technology or AI’s contribution. Because we talk a lot about, like people talk a lot about like how companies are trying to measure AI’s like actually, ⁓ you know, contributions to the company right now, the profitability. How do we actually understand AI’s contributions to the economy? I guess it’s two separate questions, but yeah, help me understand that.Tom Nunlist (22:19)Yeah, this is a really interesting question. And I think frankly, it’s one that the government is just trying to figure out itself. For years, of course, it was just GDP. So you win if you bring your area GDP, which is great for encouraging growth until it encourages the wrong kind of growth or encourages the wrong kinds of projects. And so I’m a little bit less familiar off the top of my head, but it’s something my colleagues have looked into. ⁓as well is how these KPIs might be changing. And again, from this shift from quantity to quality, I think at the end of the day, probably something like GDP is simply the easiest thing to sort of see. But certainly, and that’s like if you’re like a mayor, I guess. But for people that maybe work within different ministries or in like...specialist areas, whether or not they do a cool project along these lines, whether or not they brought, they fostered the emergence of a new giant in their district. That’ll be looked on favorably. So in terms of who actually sets these KPIs, I think that would actually go down to the personnel department ⁓ and how they interact, how the personnel department decidesthings to include on there, some of which will be from NDRC’s AI plan and some of which will be from like totally other different things. I can’t tell you what their score rubric looks like. But again, the message here of going this like broad top-down kind of thing, what officials will be doing is, youlooking at the communication of these targets, right? Looking at the messaging and interpreting them for their district, right? So what do I need to do to make that happen here? And that’s the way forward for my career, right? And also to connect this with what you were just talking about in terms of local specialization, right?what’s going on in Harbin, the local conditions there are different from in Shanghai or in Hangzhou. And so I think in the ideal way, and the government uses this phrasing a lot, is to have things definitely specially adapted for your local conditions, right? Don’t just do exactly what we’re saying, like make it work for you.Right. And so in the ideal world, you would have like different things going on everywhere and they would all be complimentary. I think what happens, what tends to happen is that you have duplicative efforts, you know, which of course we see, you know, everyone’s talking about now in the auto industry. my gosh, there’s a hundred auto companies and they’re all, you know, in a giant battle Royale that is destroying value, you know, rather than. Yeah.Grace Shao (25:06)Yeah, the price war right now. Yeah.Actually, how do we understand this? think because for the sake of, know, understanding Chinese policymaking for say, Western investors or Western companies, like say, MNCs operating in China, and in the day is to help thembetter their operations, right? So then how do we understand this from that perspective? Say your client’s M &C and they’re saying, seeing, okay, AI plus is being rolled out on a central level. Then they are like, how do they decide? I don’t know where to put their plan, to build out their operations. How do they kind of make that judgment comparing provinces to provinces? And I think to your point, you kind of have...answer this in the sense of like maybe if you’re industrial machinery you go to Harbin right but if you’re consumer goods you’re Shanghai but are there any other things that companies need to be aware of or investors investing in companies are coming out of these different problems should be aware of?Tom Nunlist (26:00)Yeah, great question. So I think probably the first choice, the first thing to look at is just, you know, where are the hubs for what you’re doing, right? If you’re an automotive company and you’re looking to make any of these, well, might go to, you might go to Enhui, right? Hefei, sorry, I forgot it for second. You might go to Hefei because that’s where a lot of the new energy vehicles are.Right, and then from there, and this I think is a bit more unique ⁓ to China, is if you’re a corporate and you’re trying to be successful here, one of the first things you need to do is align with whatever the government is trying to do. You know, that doesn’t mean do exactly what the government asks you, right? But you know, figure out what officials there want, what their KPIs are, what their existing programs are, and how do you align your corporate goals with that?⁓ And that’s how you get support. That’s how you get buy-in. That’s how you’re ultimately successful, right? You know, as in sure it’s no secret to anyone, you know, the Chinese government just has a much bigger voice in the direction that the economy is going, right? And the things that are happening in the economy and, you know, companies and investors absolutely, you know, have to listen to what that voice is saying.I think for investors as well. So where are these companies collected? Where are the big hubs for the industry that we’re investing in? And also, what is the government itself saying that it wants? And which companies do we think can...Obviously, of course, first deliver on the market promise, like do what they’re saying you’re trying to do, but are there opportunities here? Will they get this kind of support from the government that is a factor that is larger here than it is in other places? Probably maybe any other place.Grace Shao (27:46)Yeah, I think it’s also like the point you’re saying, it’s not really like you have to do what the government says, but it’s like you might as well lean into, like, I guess, lean into it, right? Like there are going to be favorable policies for your industry, certain areas, municipal areas, you might as well lean into it to optimize or to like maximize your success rate or your success possibility, right? So on thatGrace Shao (28:10)point actually, I’ve heard that there are quite a few AI pilot zones. Like, you know, right now, I think for the West, people only know about Shenzhen, Hangzhou being kind of the tech innovation centers, obviously Beijing, Shanghai playing a big role for corporate headquarters and obviously where investors sit, policymakers sit. What are some other major cities that are actually quite relevant to this like AI growth right now or are considered AI pilot zones?Tom Nunlist (28:35)think those would honestly be the main ones. know, Shanghai, Beijing, you said, Shenzhen, Guangzhou, Hangzhou, like these are the places where, you know, a lot like the most action is happening, right? Especially in an area where we’re talking about, I mean, it depends on what we mean, right? So like if we’re talking about just raw AI development, making new LLMs and stuff like that, you know, one of the big, you know, stories is that there’s only so much talent out there that can do that.⁓ and this talent will gravitate towards some center. And there’s only a few of those, only, not everyone can have those people. Not everyone, those people won’t go everywhere.Right, AI, but back to what AI Plus is about, right? AI Plus, all of these other things, right? And having that in various sectors, I think where other cities will excel or have the opportunity to excel is where those hubs are, right? So if we’re trying to add AI into auto manufacturing, that’s gonna happen in an auto manufacturing hub.Right. And I think that actually speaks to the important thing that folks need to be looking out for. You know, at this point, know, we’re, you of course, at the high level, you know, we’re talking about sectors. OK, we AI in the research sector or want AI in the health care sector. But I think what’s most important is going to be looking out for not which sectors it revolutionizes, but which specific use cases, right, are going to be.Grace Shao (29:59)Mm-mm, I see.Tom Nunlist (30:06)most obvious to implement.Grace Shao (30:08)And actually on that point, which use cases, let’s put it that way instead of sectors, do you think are kind of showing the most promising mass consumer adoption of AI, gen AI as we know it? So I’m not talking about like the buildup of LLMs and everything. I’m saying, know, when DPC came out, there was a media frenzy of stories about how China’s like home appliances are even adopting AI, EVs are trying to adopt AI, you know.Tom Nunlist (30:13)Yeah.Yeah.Grace Shao (30:34)I mean, obviously that kind of hype has gone, like, moved past us, but like, in terms of whether you want to use sectors or use cases, where do you see actually China right now really leading in adoption? And where do you think we’re seeing the trend going towards maybe in the next three to five years?Tom Nunlist (30:50)Yeah, think ⁓ it will continue to penetrate more ⁓ on the consumer side, just on of like AI services that are available to everyone. mean, that’s sort of the biggest thing right now. Whether or not we can get consumers to pay in China, I think is a little bit different of a question. But in terms of specific areas, I think it’ll be where we’ve already seen AI ⁓ have quite a bit of traction. So in like logistics and transportation where, you know,with like self-driving is kind of almost here and you know we have the the nev is this it’s a software-defined vehicle and we’re going to be like a ready integration for ai into the features of the vehicle that’ll definitely be one you know another one thinking about ⁓ that comes to mind is is agriculture which i you know ⁓ i can’t name a specific company or or a project but ⁓you know, drones are becoming ⁓ large and, you know, helping to manage big farms, like do things like crop spraying, you know, or inspecting or like, also not just in agriculture, in inspecting power lines, drones are not used to do that. It’s actually physically hard to get up there, right? And so there’s AI use cases for that, right? It can go into like visually inspecting, right? Or visually help, you know, irrigate your crops and so on and so forth.So it’ll be things like that, right? Where we’ve already started to see new things happen, AI being used a bit. And now these new tools and the growing power of these tools will enable it to really actually happen.Grace Shao (32:28)Yeah, definitely. think like, when I first saw and tried out a few of the EV cars, this is even like during COVID, this is like three, four years ago, I was shocked by I wouldn’t say they’re like genuine power, but how tech savvy they already are. had voice control, each of them already had a built in robot, you can control your like windows, you control your heat, like the heat of your seats by voice recognition, voice control. And I think like you said,Tom Nunlist (32:42)yeah.Grace Shao (32:52)implementing GEN.AI into it just means that it can actually embolden it more, right? Do more things or right. So that’s really interesting. I think I want to double click on one question that a lot of people are kind of debating. know, China’s approach innovation often is said to be, you know, innovate and then regulation comes later. Europe obviously takes another extreme case of like hyper or not hyper, but like a lot more.Tom Nunlist (32:57)Yes.Grace Shao (33:16)cautious and safety, you know, safety cautious and like, you know, regulation comes first. And some people are complaining about how it’s hindering innovation or innovation going into production. Right. So I guess my question right now is you’ve been in the AI safety and policy space for a long time now in China. Do you think that actually you must give up safety for innovation or are there other ways that you’re seeing people actually being able to have safety andinnovation co-exist and co-develop and maybe taking data privacy as an example or how did Deepsea come through if there was so or let’s just talk about that space.Tom Nunlist (33:50)youYeah, this is an excellent question. And frankly, I think one of the most underappreciated or even like misunderstood aspects of the AI story as it stands right now in China. I mean, there was a point not too long ago before the EU AI Act, which you mentioned where, you know, China had, you know, the strictest AI regulations on the books in the world. And yet, you know, DeepSeq was still clearly able to emerge here and,you know, become what it is, right? And I think the story here is that, you know, China is, I think, as most people will understand, a very security conscious, you know, country, but it is also highly flexible, right? And the interesting thing, the sort of interesting story, like when ChatTPT first came out, there was this mad scramble.among regulators to get a handle on it, right? Because it was gonna flood the internet, you know, with these tools and man, what are the impacts gonna be, you know, like just a real sense of urgency to try to like write something immediately. And so there was a period of about a year and a half where you had regulation after regulation and, you know, they...you know, if you looked at them in line, you’d think they were different, but they were actually kind of rewriting one another, and it was like all very ⁓ messy and a very confusing space. But then, you know, China was able to kind of like find what its bottom line is.and then be flexible and adapt from there, right? So it was, you know, hurry up, let’s do something. Let’s kind of see what’s gonna work, where it might be too far, and then dynamically kind of like dial back.So one of the interesting, I think probably the most interesting single event of this story was there was a registration system that was created where if you want to publicly release an AI tool like a chat TPT, it has to be registered with the state and blah, blah. And then some requirements started to be built on top of that. And there was a draft that said at one point,all of your data that you need to train your LLM with needs to be verified as true. And the AI research community came back and said that this is impossible. Like if this is implemented, will, know, progress will grind to a halt. There is no way we can do this, right? There was no official response to that, but the final version of the rule did not contain that.Right, was that was walked back. was an idea that was tried out, that was an explored, you know, and eventually, you know, was abandoned because it didn’t work. And so I think, you know, one of the sort of like, again, underappreciated or even unknown strengths of China’s regulatory system is that it can be flexible in that way.Grace Shao (36:27)Right.Tom Nunlist (36:46)in an ungenerous interpretation of this, which you hear from a lot of foreign companies and rightly so because there are drawbacks of this, is that regulation can kind of seem all over the place and arbitrary and you never know what things are gonna change next.And certainly in emerging areas, that is true and very challenging, you know, if you’re in a corporate compliance type situation. But the plus of that is it can be, you know, quite flexible and adapt to, you know, what the perceptions of the needs are kind of as they’re coming up, which in, you know, an environment as fast developing as this one, where again, new problems might emerge tomorrow. I think that’s a really important strength or really useful strength to have.Grace Shao (37:29)We could be quite reactive in the sense that they would actually react to what the industry and the actual practitioners at the leading frontier, technology development, want or need, right? To really help and regulate the technology, yet also not hinder any progress. I think that’s really interesting and it’s a very fresh take on it. I haven’t really heard that before, but I think it’sit makes sense. And it also kind of explains what you said about some people’s kind of complained or misunderstanding of this whole like murkiness. so you said that the AI Plus initiative really it’s been around, like not been around, but like the AI policy or the plan has been around or the idea has been around. And then there was a 2017 National General AI Plan as well,There’s also the made in China 2025 plan, all these big grandiose plans that have been really pushing forward AI or robotics and just technology development in general. as you said, policymakers and regulators can actually be quite reactive. So over the last, I guess, 10 years as these three mega plans been rolled out.How have you seen these things change or how has the policy makers really a change in terms of their sentiment or the attitude towards this technology?Tom Nunlist (38:40)Let’s say the biggest thing, so taking a bit of a longer view, so science, tech, and manufacturing development has been a priority of the governments for a very long time, since the late 90s. It’s been kind of on this top priority list. And so one shift I’ve seen in the past few years is side tech development moving from one of the list of important things into the top thing.like the most important thing. It’s like that is ⁓ kind of an organizational principle, right? Or like a driving organizer of the whole party, right? And again, that’s because of the perception of what the state’s needs are at this point. In the past, in the sort of like last formulation, right? Of like what the country needed, right? It just needed growth.It’s like, it’s the late 70s, we’re into the 90s and 2000s. We need to just grow. We need more people and jobs, we need production. That’s what we need. Now that’s not what they need. We need quality growth, we need to move up the value chain, we need to avoid the middle income trap if we can, expand people’s incomes, become a more efficient and a more technologically driven society. And so the sort of prioritization,and some of the character of these plans have changed sort of in line with that. Some other things I think have stayed the same or strengthened rather, right? So with Made in China 2025, which this not really talked about explicitly anymore because of the political sensitivities it creates in the US, right? But the sort of view, right, was that Chinayou know, doesn’t want to be vulnerable, basically to, you know, always reliant on outside technology and wants, you know, these things for its own, right? Wants them to be secure and controllable. It wants to have, have its own thing, right? That of course, I think, you know,in light of the subsequent ⁓ US effort to strangle the semiconductor sector in China is even more of a priority. So it’s not just move the value chain and get incomes up, it’s also create these fundamental technologies which we absolutely cannot have as a vulnerability.Grace Shao (41:03)Essentially kind of push more honed in on the self-reliant focus than they previously didn’t really have to, right? It was also kind of a reaction as well. Okay, I think I want to go in some quick questions. ⁓ You did answer a bit of it, but one overhyped and one underhyped province or city that you think people are not noticing enough outside of China.Tom Nunlist (41:28)Yeah, again, would say,Yeah, they’re not really as specific over under Hype City that I can think of. But yeah, I would say go back, double down on the point of like, you know, look at where different specific hubs are, right? So right now, you know, especially the US talking about AI development sort of in general, right? Like the rush to AGI, you know, so on and so forth, right? The AI plus plan is about doing things in the real world, right? So I think where a lot of like really fascinating stuff is going to happen is where those real worldthings are in China, right? So like where we have many filtering use cases actually emerge and that’s going to be sort of all over the country.Grace Shao (42:02)Right.Right, like CN maybe for renewables, but like Hefei for you say auto, and then like even like Baoding and Hebei for like auto. You get at least like second, third year cities that are just like actually relevant, but only if you’re in the sector, you would know, right? And that’s a really interesting take. So what is one metric that a policy analyst like yourself should be really tracking or focusing more on?instead of just, you know, maybe what we’re seeing on the headline is like, you know, this crazy chase for like benchmark frontier technology, frontier of LLM benchmarking. How do we actually track or judge real AI diffusion in the economy?Tom Nunlist (42:48)would say it’s probably more along the lines of traditional measures, So penetration, productivity, profitability, wage and efficiency growth. Again, the emergence of those scenarios, Are people actually out there using it in the real world? So I think it’s look for those traditional.tangible things, right? Again, I mentioned that Chinese consumers tend to not want to pay for consumer-grade AI tools. If that’s something that changes, right? If they’re good enough where people are willing to pay for it, wow, maybe that would be an enormous indicator.Grace Shao (43:16)Yeah.I don’t think anyone’s gonna want to pay for like, you know, consumer app. The culture, right? Like no one wants to pay. I don’t know, I switched my brain on and off when I use like Western apps versus Chinese apps. And when I’m on a Chinese app, they pop up, they’re like, pay for premium. I don’t want that filter anymore. I don’t need this sticker anymore. I’m like, I’m not paying. You just have a different mentality, right? Because you do get too many goodies for free already. It’s very...Tom Nunlist (43:27)I think he wants to go to pay for a I know.Yeah.Yeah.Grace Shao (43:51)It’s very hard, think. The barrier is very high. The threshold. I have one last question for you. And it’s a question I ask everyone that comes on to differentiate understanding, which is what is an unconventional view you hold? And this could be about work or something in life, you know? But what is something that you think about and you’re like, oh, maybe I don’t say this out loud, or maybe this is quite different from what my peers think?Tom Nunlist (44:13)What was an unconventional view I hold?I’ll go one with topic specific here. That’s because it’s come up recently in ⁓ fights I get into, Twitter fights I get into with people. There is this interesting and I think not totally off the mark concern with AI that it’s gonna basically make us all dumber. Students are gonna outsource learning to AI. There was a case study that did the rounds about doctors using AI tools to help them spotcertain types of cancer got worse at it, know, like after relying on the tool. And that’s a real concern. I think it’s something that, you know, there’s some red flags that seem to say that that might actually be happening. But I think the real problem might be a bit more nuanced than that. think it might, my hypothesis is that it will create something like a ⁓ skills or performance gap.between different parts of the population and exaggerate it. So, whereas some groups of people might become reliant on it and become de-skilled in their job, definitely. And then in some cases, that might be what we wanna happen. I we don’t want everybody, I mean, that’s sort of the promise, not have to do certain boring things. But I think for a smaller portion of the population, it is gonna be a massive learning and development.accelerator, right, to really help you to get good and improve. And so, you know, I mean, beyond, you know, whether or not I’m right, I don’t know I’m right, it’s just a bit of a guess. You know, I’m wondering where that gap kind of might be and how large it’ll be, right? So is it going to be 90 % of people get dumber and 10 % of people become super learners? You know, or is it, you know, somewhere in between?That’s my unconventional view. It’s gonna create a skills disparity.Grace Shao (46:04)Yeah, I actually kind of agree with I think it would make people who are relying on it for skill set like vocational skill almost like just the art of, know, not art, but the skill or ability to write a press release or draft a basic news piece or you know, build a DCF model or you know, do some quick basic research that might become dumber in the sense that you don’t know how todo it in a traditional way. But I think the arguments also like say 34 years ago, people are like, you have the internet now. You don’t even know how to use a library anymore, which I think our generation honestly, I don’t really know how to use the library very well. Like I go on my loss. I don’t know how, you know, how to find books essentially from alphabetical order and, you know, like finding the topics, but we do learn how to find more information in some sense, right?But I think to your point of like, ⁓ it will help people learn a lot faster, but it will require a new kind of skillset, is like, you can access all this information, can you decipher it? Can you dissect it? Can you actually pick out what is correct? What is actually relevant? Because there’s so much noise and clutter, which is kind of similar again to our generation where we had to use the internet to find information, Versus like our parents generation had to like walk into the library and just like.Grace Shao (47:19)go through like 10 books, right? ⁓ But that’s super interesting. Thank you, Tom. Really, really appreciate your time. This was super insightful. It was really helpful for me to even learn about how to understand how policy was made in China, how it might affect businesses and investors. And yeah, this was just super insightful and a lovely conversation.Tom Nunlist (47:21)Yeah. Yeah.Yeah, thank you, really. It was really lovely to be on the pod.AI Proem is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to AI Proem at aiproem.substack.com/subscribe

Joining me today is Natalia Cote-Munoz, a policy strategist, writer, and International Strategy Forum (ISF) Fellow— a program by Schmidt Futures that supports rising leaders at the intersection of geopolitics, technology, and public service. Natalia has served in the U.S. State Department, leading foreign policy think tanks and crisis diplomacy roles. She is a graduate of the Harvard Kennedy School and speaks English, Spanish, and Mandarin, among other languages.In this conversation, we discuss Natalia’s unique upbringing as a third-culture kid, her experiences in tech diplomacy, and the evolution of US-China relations in the tech sector. Natalia reflects on her recent return to Beijing after a decade, sharing insights on the rapid technological advancements in China, particularly in AI and digital payments. We also discuss her observations of how diplomats are trained as an international relations teacher at China’ Foreign Affairs University, how AI cannot be replacing humans in diplomacy, her embrace of AI in productivity and creativity work while she was experiencing a concussion, and lastly her unconventional belief about the societal views on Labubus, highlighting the cultural differences in perceptions of childishness and professionalism.In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.Chapters01:30 A Third Culture Kid’s Journey04:21 Evolution of US Tech Diplomacy11:32 Reflections on Beijing After a Decade17:54 Exploring the Red App and AI Conversations - Doomism vs. Optimism28:35 Education and Talent Development in AI38:20 Exploring Student Aspirations in International Affairs40:33 The Role of International Faculty in Education41:56 STEM vs. Liberal Arts: Educational Mindsets47:41 AI as a Productivity Partner: A Personal Journey56:24 AI in Diplomacy: The Human Element01:01:26 Legitimacy in AI: Who Builds It Matters?01:02:00 Cultural Perspectives on ProfessionalismAI Proem is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to AI Proem at aiproem.substack.com/subscribe

Kevin is an investor at a family office, where he leads AI investments across asset classes. His career has spanned roles as a venture capitalist, startup founder, and software engineer, with experience in both Silicon Valley and New York, before moving to Asia. He brings deep technical and product expertise across domains from machine learning to enterprise software. In his spare time, Kevin writes East Wind, which is focused on technology investing.In this conversation, Kevin Zhang shares his insights on the evolving landscape of AI investments, the implications of hyper-scaler capital expenditures, and the future of AI model training. He discusses the cultural differences between investment ecosystems in the US and China, the valuation of private market companies, and the role of neoclouds in the AI sector. Kevin emphasizes the importance of capital and distribution in determining the success of AI companies and reflects on the future of work in the context of AI adoption.In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.Chapters00:00 Kevin Zhang's Journey From Software Engineering to VC to Equity Investment02:04 The Hyper-Scaler Capex Debate04:31 The Capital-Intensive Nature of AI Models07:49 Future of AI Capex and Market Dynamics11:43 Understanding Private Market Valuations14:49 Consensus Capital and Investment Strategies17:10 Cultural Differences in Investment Ecosystems21:50 The Future of Chinese AI Companies23:50 Capital and Distribution in AI27:47 Open Source vs. Closed Source Models32:22 The Role of Neoclouds in AI40:40 Investment Opportunities in AI and BeyondTranscript generated by AIGrace Shao (00:01)Hi everyone, this is Grace Shao. Joining me today is Kevin Zhang. Kevin is an investor at a family office where he leads AI investments across asset classes. His career has spanned roles as a venture capitalist, startup founder, and software engineer, with experience in both Silicon Valley and New York, before moving to Asia. So he’s now based in Asia. He brings deep technical and product expertise across domains from machine learning to enterprise software. And in his spare time, Kevin writes a blog on Substack called East Wind. Go check it out.It’s focused on technology investing. So Kevin, thanks so much for joining us today.Kevin Zhang (00:33)Hi, great pleasure to be here.Grace Shao (00:36)Yeah, tell us about yourself. You’ve had quite a journey, you know, from Silicon Valley and now into your base in greater China. You’ve worked as an engineer and now an investor. You know, that’s quite unique. Tell us about your professional journey.Kevin Zhang (00:50)Great. So I guess going all the way back to my college days, ⁓ studied computer engineering, both in Canada and then states for grad school. And then spent most of my career in the States ⁓ in Silicon Valley and New York. So started as an engineer at a company called Salesforce. So they make CRM software before ⁓ transitioning to a couple of venture funds, ⁓ one in the Bay Area, one in New York. So primarily the focus has been onearly stage software, AI investing. And in these days, I look primarily at public equities, specifically focusing on the companies that have exposure to AI or ⁓ businesses who ⁓ will see a re-acceleration of growth because of AI.Grace Shao (01:39)Awesome, so we’re gonna go straight into this. I think we have a lot to cover. ⁓ First, let’s get into the hyper scalar capex kind of story. We’ve been seeing jaw dropping capex commitments, alphabet raising, AI capex too, 85 billion, meta committing hundreds of billions, ba-ba 10 cent I think over the next couple years, also committing like 300 billion upwards. It’s just these crazy numbers.Do you see this as a necessary investment or do you think they’re absolutely overspending right now and they’re creating a bubble?Kevin Zhang (02:11)I think if we look at past ⁓ technology cycles, there will always be, or there has always been ⁓ some overspend across the industry. ⁓ So I think ⁓ bring it down to on a per company level, I think things get more nuanced, right? So looking at, for example, alphabet, right? How much of that is internal workloads for search, right? For their Gemini answers, ⁓ looking at open AI where how, ⁓whatever their, can we maybe start this again? Let’s do a rewind. Okay, great. ⁓ So I think across the industry, if we look at past ⁓ technology waves, ⁓ most famously in the ⁓ initial tech kind of a dot-com bubble, the industry has overspent, right? But as we zoom out, ⁓Grace Shao (02:49)Just go ahead, just restart your answer.Kevin Zhang (03:11)the spend becomes more normalized and then the demand ⁓ ends up catching up. And so I think the real question is on a per company basis, right? Whether it’s Alphabet, Meta, Azure, with Microsoft, AWS, how much of that is overspending? How much demand can they generate, right? With their ⁓ whale customers? And then if they end up overspending, how many of these players can survive, right? So for some of the smaller players,who are comparatively more leveraged, who don’t have the cash flows to support some of the CapEx, right? They’re maybe in a little bit more of a dangerous territory than one of the Meg-7, where they’re generating a significant cash flow to ⁓ fund their operations.Grace Shao (04:01)And I think this kind of goes into also one of your writings recently. You were saying that, like, look, it’s all these big tech that are able to afford ⁓ spending on, like, LLM training and inference, as well as whatever infrastructure that’s needed to really build out, like, sophisticated LLMs. And you said it’s basically foundation models are a rich company’s game. Why is it the model layer is so capital intensive? And do you think that means we’re going to see the startups just kind ofin this field kind of just die out one by one or acquired or what’s trajectory going forward?Kevin Zhang (04:38)I think there are two paths that the industry can take. So the default path is if we look at ⁓ the progression of ⁓ costs for model training, whether it’s pre-training or post-training, ⁓ each generation has been significantly more expensive. So ⁓ many years ago, it might have been ⁓ several hundred thousand dollars. Then it went to the millions, tens of millions, hundreds of millions to train a model.So assuming that trend holds, ⁓ we’ll see kind of billion dollar training runs, right? So taking a billion dollars to train, let’s say GPT-6 or GPT-7. And so if that is kind of where the world goes, then these companies will need to raise more and more capital to fund their training. They will raise more and more capital to fund their inference, right? So once you train a model, how do you serve it to...⁓ and users, right? That’s also very, very expensive. ⁓ However, if you’re of the view that there will be, or transformers, which is kind of the models that are used in ⁓ LLM or ⁓ used in things like ChatGPT these days, ⁓ if you believe that there will be other model architecture paradigms ⁓ that are going to be significantly cheaper, then maybe there will be another kind of startup that comes and disrupts.the entire business model of an OpenAI or Anthropic or any of these labs.Grace Shao (06:07)and was DeepSeat one of those that kind of disrupted the whole model.Kevin Zhang (06:11)⁓ I think not necessarily. think ⁓ going a little bit into the weeds, the, I think, five, six million dollar ⁓ final training run touted in their paper, ⁓ that was only for the final training room, which is not inclusive of the GPUs that they’ve acquired, ⁓ their human capital.all the prior training runs and experiments that they’ve run. And then also within kind of AI training, if you basically train last year’s model today, right, it’s significantly cheaper than if you want to train a frontier model. ⁓ And I think Anthropic had a paper where, or had a blog post where basically ⁓ for if they want to train a similarly similar model to the DeepSeq R1 a year ago, it would have been 10x more.Right, so essentially the costs track. And so it’s less so disruptive ⁓ than I think some of the folks in media might have thought. And certainly they have ⁓ made certain architectural improvements as well as inference improvements at DeepSeek.Grace Shao (07:32)So like if that’s the thinking, like looking ahead next three to five years or even a longer run, is this like the hyperscaler arms, I guess, if you put it that way, is it gonna just keep on climbing up that capex or will it eventually plateau or is this question we will never know.Kevin Zhang (07:49)⁓ it depends on like, like how long the labs, right? The, the open AI anthropics, even Google can sustain, ⁓ this pace of model improvements, right? So we we’ve seen a little bit of plateauing in, in, in the past year or so, ⁓ as, as we’ve kind of reached the limits of pre-training, right? So, ⁓ right now, a lot of the emphasis is on like these like thinking models, right? So, ⁓ when you actually type a prompt,into ChatGP, you know, think for a while and then that tends to ⁓ generate better answers for you. ⁓ And so it feels like the like assuming compute requirements continue to wait, what can we start the question again? What was the specific question? Like what is the arm? ⁓ Okay.Grace Shao (08:43)Yeah, I’ll just redo it.So if we use this logic, right? Like, does that mean that in the next three to five years, we will in the longer run that the capex numbers will just continue to climb? Or eventually we will see this hyperscaler arms race kind of plateau out a little bit because I don’t know the day, like these are already like crazy numbers, right? Like we’re looking at a couple hundred billion dollars put into training in the next three to five years. That’s the plan. But how does one keep up with this kind of money?Kevin Zhang (09:13)Okay, so there’s two things, right? There’s training and inference. ⁓ And so on the training set of things, the assumption is ⁓ labs will continue to require more and more compute to train more and more expensive models, right? So let’s say the next model takes 10 billion to train, right? And the model after that takes 50 billion to train. Then theoretically on the training side, that tracks. And then I think where a lot of this capex is going, especially if you look at⁓ the Meg 7 where they’re putting 70 to $100 billion ⁓ per year per company. I think a lot of that is the expectation of inference demand. So as you put these models to production, whether it’s large language models, whether it’s recommender systems, image models, video models, that demand will catch up. So as it stands, there’s a mismatch between ⁓ the capital outlayinto these data centers versus the revenues that Gen.ai companies are generating. So if we look at OpenAI, Anthropic, they are the primary beneficiaries in terms of how fast their revenues have grown and the absolute scale or the relative scale of their revenues relative to even companies like Cursor, who grew very, very quickly to 500 million in ARR.And so ⁓ in terms of like software revenues, we’re kind of in the tens of billions range, whereas ⁓ for ⁓ hardware CapEx or data center CapEx, we’re in the hundreds of billions, right? So assuming that ⁓ software revenues, two to three X year over year, then eventually it will catch up to CapEx if end users.⁓ enterprise customers find that they’re not getting ROI from these Gen.ai ⁓ apps, then I think that’s where the house of cards ⁓ collapses.Grace Shao (11:14)Yeah, actually, let’s just like talk about the private market valuation quickly. Like right now, OpenAI is valued at like over 500 billion, something like 180 billion, right? Like startups like Hercer, Lovable, Chasing Billions, or what they’re calling trillion dollar ambitions right now. ⁓ I think the Lovable CEO said they want to become the first trillion dollar business in Europe, right? How should we make sense of these numbers? Like, I’m not a quant person. They just sound like humongous numbers.Can you explain this to us, like how to make sense of this? these are just, ⁓ does it make sense for these companies to be valued at this high in the private market right now?Kevin Zhang (11:52)Yeah, so I think that’s a really good question. ⁓ So the ambition for a frontier lab like OpenAI Anthropic is to be one of the big boys at some point in the future. And so taking ⁓ OpenAI as an example, right? So if an investor is of the belief that they will eventually build their own cloud, they will get into robotics, ⁓ their ⁓ core lines of business, right? Chatchi PD as well as their APIs becomereally large businesses, right? So let’s say TriGPT is embedded in various enterprise ⁓ kind of customers. And if they’re actually able to charge, let’s say two, 300 bucks a month, right? As ⁓ more and more white collar workers are reliant on open-air technology, that tracks to a market, even in the enterprise side of things, that’s several hundred billion dollars, right? And then if you’re also of the belief thatGoogle search will be disrupted and Gemini somehow fails to catch up to OpenAI, then they could also run ads on the consumer side of things. So once you add all these kinds of lines of businesses together, an ⁓ optimistic person might see kind of a line of sight for OpenAI to be this like three to five trillion dollar company that some of the Megs have been at already.are an investor at the $500 billion ⁓ mark, then I think that’s the return profile that you’re looking at ⁓ before kind of taking into account all the dilution from subsequent funding rounds, stock options, et cetera. And then moving down to the application layer, I think these companies are making is if we’re able to replace broad swaths of labor,⁓ and you are able to command pricing that’s at some proportion of the ROI that you deliver relative to just replacing like a human headcount, then the exit values for these become enormous, right? So then cognition at 10 billion might sound really reasonable. I think as the issue right now is the exits will be very spiky, meaning we’ll see a lot of zeros.⁓ and you’re going to see a lot of companies really become those $10,000,000,000 companies. And then for a VC fund, you have limited shots on goal. And so as the entry valuations ⁓ increase, you have less shots on goal. And so on a per investment basis, ⁓ your risk ⁓ increases quite a bit.Grace Shao (14:40)So in that sense, you don’t think we’re nearing the ceiling of model layer evaluations or anything. We haven’t hit the peak of the bubble or anything yet.Kevin Zhang (14:49)⁓ The markets are definitely frothy, but the winner will be much bigger than we, ⁓ I think, originally estimated. And so if you are one of those investors that are in these assets, I think you’re going to be fine. If you’re not, then I think, which is probably the majority of these funds, ⁓ I think they’re going to be hurt.Grace Shao (15:10)So in your writing on Substack, you’ve argued that, you know, consensus capital is crowding into foundation models and fra robotics, I think. But, you know, are there areas that you think are under invested and still in the private market? Like, where do you see, like, overlooked opportunities right now?Kevin Zhang (15:26)I think it’s less so ⁓ maybe overlooked opportunities in AI, right? So like a generalist VC is able to allocate capital across different things, right? So that could be psychedelics, that could be robotics, that could be biotech. so, or consumer as we’ve talked about before. ⁓ so figuring out kind of what the market dynamics are for those industries whereyou’re just non consensus enough to be that first check in, but you’re consensus enough that at the next round, ⁓ whatever you’ve invested becomes consensus. And this was like the subject of some Twitter debate ⁓ with Martin at Andreessen where ⁓ he was arguing it’s not bad to invest in consensus deals because like in the end, like some of these deals ⁓end up generating huge returns. And we know that even with an AI, like if you’re in a consensus bet that ⁓ pans out, assuming OpenAI is that company, then you’re still seeing like a 10x gross return, right? Assuming one of these companies becomes like five trillion.Grace Shao (16:43)Yeah, because consensus, guess, it’s for a reason, right? I was speaking to a few VC investors in the Bay Area a couple of weeks ago, and they were saying, like, some of them are kind of complaining that their bosses are just chasing logos rather than the differentiated bets. But I think in some ways, like you mentioned, if it’s an open AI and it’s still going to be the market leader, market winner, you’re still going to come out on top, I guess. ⁓ Yeah. I want to hear, OK, taking a step back from...Kevin Zhang (17:05)Yep.Grace Shao (17:10)these questions, think from a cultural perspective, you you worked in Silicon Valley, New York, and now like, you you moved around in greater Asia, greater China. ⁓ What do you think, like differentiates the two ecosystems the most in terms of like the investment space and then maybe even just like some kind of high level work, cultural tech tech space observations?Kevin Zhang (17:33)Yeah, I mean, there’s a couple of things, right? So one is the abundance versus the scarcity of capital. ⁓ And so ⁓ in the US, there’s still a relative abundance of capital where ⁓ as an entrepreneur, it’s comparatively easy, easier to be funded versus ⁓ a similar entrepreneur in Europe or Asia. And so and the other thing is,you have capital at every stage, right, from seed through growth. ⁓ And so the market as a whole has more shots on goal, more opportunity to experiment versus China, right, where ⁓ there is a comparatively ⁓ or significantly less capital, especially US dollar funds in the past couple of years. And so on the investor side of things, ⁓ they arealso more risk off, right? Because for some of these ⁓ Chinese VCs, they might not be able to raise another fund, right? So each shot on goal ⁓ is a very heavy bet, right? Versus entries, and if you deploy like a fund very quickly, could probably just raise another one very quickly as well. ⁓ And so ⁓ if, you know, entrepreneurs can be more risk on in the US, investors can be more risk on thanone of these bets will pan out and then that becomes the next big company versus in China where you could raise less capital at lower valuations, less capital at growth. ⁓ And I think where the domestic VCs might be ⁓ extremely careful and maybe not having the same kind of venture parallel mindset. I think that’s...that reflects on the products that you can build and the scope of ambition for entrepreneurs. And I think broadly this is why more more companies are trying to do the true high model, right? Where they might raise their first round of funding ⁓ in China, but then quickly pivot to a Singapore or Canada, the United States, right? And then raising capital in the US.Grace Shao (19:54)Do think it’s like really affected the dynamics between investors and founders as well? Or do you think that relationship actually is still quite similar?Kevin Zhang (20:02)⁓ For the US, I think it’s still an entrepreneur’s market for the best entrepreneurs. ⁓ Like given the abundance of capital, that has not really translated to kind of a linear increase in top founders. think capital is still chasing founders every year ortaking a step back, there’s a limited number of great founders per year that can build these generational companies. And so when AI investing becomes consensus and these founders ⁓ tend to be in the US, then you have quite a bit of capital chasing these select founders. ⁓ And then within China, like there’s probably higher pricing ⁓ orthere’s more power on the buy side, right? Where if you’re one of the 10, 20 funds that still have dry powder ⁓ at the early stages or one of the five to 10 funds at the growth stages, then comparatively it’s less competitive ⁓ for the investor versus the US.Grace Shao (21:18)How do you think that’s affected, I guess, this generation of AI entrepreneurs coming out of China? you know, like actually, if you look at the six dragons or four tigers, whatever you want to call them these days, ⁓ you know, like they’re definitely nowhere as like valued as high as the American counterparts. ⁓ There are a lot of them actually finding funding from the BATs instead of, you VCs. So like has this affected theirfuture trajectory or the entrepreneur’s mindset or their business model.Kevin Zhang (21:50)Yeah, I mean, I think it’s ⁓ all TBD based on or TBD, like given the pace of innovation, right? So to our earlier conversation, like can one of these labs build ⁓ something that’s post transformer, right? Or ⁓ can one of these win in other modalities, right? Whether it’s video or image or something else. ⁓ But from a capital perspective,it’s really hard to raise like another 2-3 billion for any of these companies. And so, like my hypothesis is ⁓ some of these companies going public is a way for them to get another turn of the card, right? So it’s like, hey, if we’re able to raise 500 million a billion ⁓ on the kind of Hong Kong stock exchange, that gives us another two to three years, right, to figure things out versus OpenEI or Anthropic where every round isheavily oversubscribed and they’re able to ⁓ basically chase everything, right? Chase capex, chase kind of their core products, ⁓ as well as some of the moon shots, With opening at going into robotics, ⁓ going into ⁓ video image, et cetera, et cetera. ⁓ Obviously, maybe not to the same level of success as some of the other kind of image players or video players.Grace Shao (23:12)When we last spoke, you kind of made this like big statement thinking that eventually these Chinese ⁓ companies will potentially be irrelevant or they’ll stay within their ecosystem, right? It’s a big statement. But I was kind of curious if you could like elaborate on that a bit more. Like, I know you’re very bullish on the American leaders right now, market leaders right now, but why can’t a deep seek or a moonshot, you know, maybedo well globally as well, or would Alibaba’s open source eventually kind of take on a leadership role?Kevin Zhang (23:50)Yeah, so it’s two things, right? I think it’s capital and it’s distribution. On the capital side, like we’ve talked about, that ⁓ model training inference follow the current trajectory, then it’s a capital game. And so the companies that are able to generate the most revenues the fastest and is able to also raise the most capital will win, right? Under this kind of paradigm. And then secondly,Even for companies like OpenAI, think they realize that there’s a limit to, for example, how much you could charge for APIs. And so they’re definitely moving into applications, whether it’s ChatGP or something else. And that’s where you ⁓ can grab higher kind of margins than purely API revenues. And then as you go into workflows, as you go into the replacement of human labor, as you...Take advantage of your initial tech advantage to embed yourself into kind of the fortune 500 clients They’re not gonna kind of rip you out versus and then swap swap potential your Chinese competitor in right so I think the speed of distribution and how fast you’re able to embed yourself into ⁓ Customer workflows where those customers have a very high likelihood to pay or a high willingness to pay I think that that’s kind of the name of the gameSo capital and then distribution.Grace Shao (25:18)Because you’re basically saying monetization still has to be on the enterprise end. What about China’s AI application on the consumer end? Do you see that being one of their advantages or something that they’ve done really well in terms of adoption rate? I deep sea hit like, not deep sea, dobao has hit more than 450 million in the MAU. Deep sea even higher than that. It’s kind of crazy in terms of scale. think about just like compared, it’s almost comparable to the leading American.⁓ applications right now.Kevin Zhang (25:49)Yeah, I think from a usage perspective, for sure. ⁓ The more interesting thing is how do these companies and these products monetize? If we look at Alibaba, Tencent, guess like Baidu, ByteDance, and then even like some of the newer players like Xiaohongshu, how do they monetize? It’s through ads, it’s through e-commerce. And so I think for these,new Chinese companies, it’s figuring out how to monetize this chat interface ⁓ catering towards a Chinese audience. And I think the company that’s able to figure it out, and it could be one of the giants, existing giants, they’re going to capture that slice of revenue. But as far as the willingness to pay for a Chinese consumer, I think that’s significantly less than ⁓an equivalent US consumer where the number of people in China who are willing to pay 20 bucks a month or 200 bucks a month for kind of the higher tier of opening, that’s gonna be limited. So you have to win on ⁓ scale, right, scale of users and you have to win on these other potentially less obvious sources of monetization.My guess is it’s gonna be advertising, it’s gonna be commerce.Grace Shao (27:17)Yeah, it’ll be like value added services or like invisible ways of making money. It’s not going to be like a subscription model or anything. I agree with that. Yeah. Let’s just actually double click on China. ⁓ know, Bill Gurley, I think has been one of the more I would say pro China, but more outspoken investors in Silicon Valley. That’s not anti China, at least. I think he recently just talked about going to China as well, doing a big trip with his daughter. ⁓ You wrote about Bill Gurley being wrong.Kevin Zhang (27:20)Yeah.Mm-hmm.Grace Shao (27:47)about China’s open source models. Can you explain to the audience what was your whole thesis and why do you think he’s not analyzing the space correctly?Kevin Zhang (27:58)Yeah, I first off, like I think where he is right are a couple points, right? So the entrepreneurs, investors, ⁓ folks in technology and finance definitely pay way more attention to the US ecosystem and learn from the US ecosystem way more than vice versa, right? And so that creates a huge blind spot for US entrepreneurs and investors. And then secondly, I think there’sQuite a vibrant and talent dense kind of ecosystem ⁓ in China here as well ⁓ I think where he is maybe not ⁓ Where he might have like missed the mark one is this kind of like grasses greener on the other side syndrome, right where a US investor might feel like like China’s like the land of opportunity because of potentially lower valuations because of⁓ very competitive entrepreneurs, ⁓ etc. Whereas, and they might have this feeling that ⁓ China is like kind of the buyer’s market, right, or the investor’s market. ⁓Grace Shao (29:12)But wait, that waswhat happened with Internet error, right? So it’s not like he has no basis with this mentality or thinking.Kevin Zhang (29:20)I think that’s partially correct. I guess the question is how does an American VC ⁓ monetize ⁓ this ⁓ insight if they are potentially unable to invest in these ⁓ underpriced but potentially competitive assets given some of the limitations for American investors? And then secondly,⁓ I think the open source closed source debate is more interesting. So I think ⁓ Bill Gurley is a proponent of open source. And so the framework is like, if you are a front to your lab and you are training models that cost ⁓ tens or hundreds of millions of dollars per training run, you have to monetize it in some way.And so if you are a lab that gives it away for free, that truly limits your ability to monetize. ⁓ And that is like very different from Facebook open sourcing their models because they have a very, different business model than some of the labs here in China. So it goes back to even opening in Anthropic where initially they have the tech edge, they have the best models.eventually folks are going to catch up. So they’re really in this race to maintain being one of the top players, right, top two to three players, and then ⁓ winning distribution, right? So if you are at this Google scale as an opening eye where you have multiple software assets that people are using daily and your model is, let’s say, top one or two or three, that becomes very, very hard to displace compared to a lab that only has a model but no product.And so I think that confluence of how do you build product on top of your model layer while you have the lead is really the name of the game versus, hey, I’m going to train a pretty good model and then I’m going to monetize via API and then just give it away for everyone else. I think that’s probably not the winning game in this era.Grace Shao (31:38)Then what’s the game that Deep Seeker moonshot or any of these, ⁓ the kit moonshot that’s behind Kimmy, what, what, what should be the game for them, I guess, for them to be able to monetize eventually.Kevin Zhang (31:51)That’s a great question. I don’t know.Grace Shao (31:53)Yeah, I think that’s why like for Baba, it made sense, right? It’s kind of like the meta llama business model because end of the day, they own distribution and they own the infrastructure. but but like it’s interesting to see because even the deep seat narrative like, he’s a billionaire, he funds himself. But then that’s actually not that much money to fund. like we just talked about capex is like this is hundreds of billions. This guy’s got a billion. Like, there’s still a pretty big gap in between that he’s not going to bankrupt himself to fund this. Right. SoKevin Zhang (31:54)Yeah.Yeah.Grace Shao (32:22)It’ll be interesting to see how he monetizes or make this into something bigger.Kevin Zhang (32:22)Yeah.Yeah,I mean, the other thing is like the market is very dynamic and for every one of these labs, you’re a one hit product away from monetizing, right? And I think it’s never a wise thing to count a player out, especially if like a founder is very good and very thoughtful. ⁓ And so if some of these like companiesend up going public and raising, let’s say, their few hundred million or like a billion, they’ll have another two, two, three years to figure out the monetization piece, right? While they try to catch up in some model modality, right? So they might, they might figure out, or mini-macs might figure out, our video models are really good and truly world-class, and we’re going to build a bunch of these workflows and we’re going to be adopted by the world, right? That could be kind of interesting as well.Grace Shao (33:22)It could be a business model that we haven’t even seen before. could be something completely innovative, right? Something completely new.Kevin Zhang (33:28)Yeah, potentially.Grace Shao (33:30)Yeah. Okay, I think I want to zoom out a little bit. When we talked, you said you are looking at Neo clouds, and this is a space where I frankly know very little about. So I want to hear from you and please do explain things in layman terms and dumb it down for me. But alongside the hyperscalers, we’re seeing Neo cloud players like Coreweave, Nebius emerge, right? Like they’re making headlines. ⁓How does the dynamic really work between the Neo clouds and our old traditional cloud players?Kevin Zhang (34:03)Yeah, mean, for some of these, so defining Neo clouds like the way I think about it is it’s GPU as a service, right? So whereas a traditional cloud provider like AWS or Azure or GCP, they might provide a bunch of services, right? So they might offer databases, they might offer compute, they might offer a bunch of these other pieces of software storage. ⁓ Neo clouds kind of simplify what they offer andThe core of what they do is they help companies with training models, doing inference on the models that they train. And so the dynamics so far has a couple of dimensions. So if we look at ⁓ the upstream, NVIDIA, so they basically supply GPUs to both the hyperscalers and the Neo clouds. Their incentive is to have or to not have the hyperscalers be that big because they know thathyperskillers like Google, like ⁓ AWS, ⁓ or Azure, they’re building their own hardware. so, NVIDIA is very incentivized to build credible competitors to hyperskillers where they reduce their customer concentration. And then for the Neo Clouds themselves, like a few of them were kind of in the crypto mining space before pivoting to kind of this like AI workflow, like AI compute.And I think the game they’re trying to win is, we have this wedge where Nvidia will give us some allocation of their ⁓ latest GPUs because of those dynamics that we talked about a couple minutes ago. And we’re going to use this as a wedge to eventually become a really big player ⁓ in the AI space and maybe ⁓ for the players with even grander ambitions.to become kind of the next hyperscaler. And so you could look at like Oracle a few years ago where their cloud business was a very, very distant forth, right? In the U.S. But because of these recent contracts ⁓ with ⁓ these larger customers like OpenAI, that they’re able to ⁓ see significant appreciation in their ⁓ market cap ⁓ and kind of this like...like exponential increase in their RPR remaining performance obligations because open eyes saying, hey, we’re, we’re, we’re willing to spend a couple hundred billion dollars on, on compute with you, right. Versus, ⁓ within Azure. So, so I think the dynamic is hyperscalers, ⁓ trying to maintain their, their lead, ⁓ while trying to build more, more and more of their hard, hardware in house. And then on the Neo cloud side, it’s, Hey, we have a wedge right with.⁓ a ⁓ demand for AI, both on the training and inference side. And we’re going to use that to grow our revenues very, very quickly, potentially take on a lot of debt, right? And then become kind of one of these large dominant players tomorrow.Grace Shao (37:12)So you said that there are essentially service providers. Could you actually elaborate a bit more on that in the sense that, ⁓ again, this is really new to me, this whole sector, but people have said, NeoClouds are a real estate business. Others saying it’s a software layer value ad service. How do we actually see this? Can you explain that to me?Kevin Zhang (37:35)Yeah, so at its core, like, Neo Clouds are just clouds with like GPUs, right? And so how do they monetize? And so there are basically three ways to monetize that we’ve been kind of diving a little bit deep into, right? So one is, hey, I am an OPENAI and I’m just gonna rent your GPUs, right? I’m gonna rent your infrastructure and I’m gonna do everything ⁓ myself, right? And so for the Neo Clouds,That’s kind of the lowest margin type business because you’re basically taking some profit ⁓ or taking some margin on top of your cost ⁓ to provide that compute, right? So part of that is your initial cost to acquire that hardware. Part of it is your kind of ⁓ ongoing electricity costs or costs to run the data center. And so that’s kind of bucket number one ofmonetization for Neo clouds and then bucket number two is kind of the managed services, right? So if you want to do, if you want to provide software for training, if you want to provide software for experiment tracking, for AB testing, things of that nature. So that gets you much closer to software margins, right? Traditional SaaS margins. And so I see more and more Neo clouds going there, right? And that BS included, uh, uh,CoreWeave as well with their acquisition of weights and biases. ⁓ And then the third part is what if we provide, excuse me, APIs as a service, right? What if we give you inference, we’ll host the models ourselves and then you just pay based on the tokens generated. And so that’s like kind of the third category. And so ⁓ how this space ends up playing out is what’s the purposeportion of revenues that these Neo clouds will be able to generate from each bucket of services and products, where if they’re able to generate more and more revenues from kind of the bucket two and three, that makes Neo clouds a much higher margin business than comparatively under differentiated, you know,GPs as a service infrastructure, although running these data centers isn’t that easy. And there’s a lot of nuance between ⁓ even some of these more leading edge cloud players.Grace Shao (40:09)I see. ⁓ I think I’m going to actually ⁓ ask you one last question on investment, is I think energy is something that people have been talking about being affected by AI, obviously. And then like you said, software businesses, you know, we’re obviously already spoke about LLMs, cetera. So what do you think is a sector that is going to be affected by AI and you’re looking at it as an investor? ⁓but are not as obvious to the public or not as obvious right now.Kevin Zhang (40:40)Yeah, I think that’s hard. ⁓ So just given ⁓ most of our efforts are on the ⁓ kind of public side of things, we are kind of looking at every layer of the sack, right? So from ⁓ semiconductors, kind of ⁓ the companies that build the underlying kind of infrastructure, right? So your transformer companies, your... ⁓power companies ⁓ to your application companies. And I think it’s more ⁓ what’s undervalued relative to kind of market consensus and, ⁓ you know, rewinding the clock back one or two months, Nebius was one of those players, right? And Oracle was one of these players. So it’s more which player within which kind of layer of the sandwich is undervalued and then how do we think aboutour risk adjusted returns, which is a little bit of a cop out answer. But our objective is to build a basket of these securities across the stack where we believe that these businesses will offer outsize returns. then going maybe a little bit deeper is for a lot of these businesses, they don’t have kind of a pure exposure to AI. Meaning even if you invest in a company likea Microsoft, right, or an Amazon. They have ⁓ their traditional minds of businesses that you have to price. And then those tend to have, especially if they’re more mature, those tend to have a slower growth rates, right? So even if your AI revenues are exploding, they might be dragged down by some of these at scale, ⁓ mature ⁓ business units ⁓ or products.And so how do we think about the blended returns for, let’s say, a ⁓ much, much more or much kind of like traditional player in the kind of transformer space?Grace Shao (42:47)All right, Kevin, ⁓ anything else you think you want to share with the audience? ⁓ Mindful time or wrapping up our episode? Is there anything you want to talk about that I did not touch on?Kevin Zhang (42:59)Sure. I think a very interesting thought experiment is how fast AI will displace work. So I think my perspective is going to be much slower than folks might think in Silicon Valley and much faster than the rest of the world thinks. Meaning next year, 90 % of the code is not going to be written byor at least like code and production is not going to be written by AI, right? Or I don’t see like broad swaths of ⁓ workers gets automated. ⁓ But I think is AI adoption going to be a 20 year kind of journey? think, no. I think ⁓ for a lot of these professions, it’s going to be this like five to 10 year disruption.And then we’re already seeing some of this ⁓ for new grad hires, right? Where a combination of AI giving more experienced workers a higher leverage, ⁓ as well as some of the broader macro headwinds ⁓ in the US ⁓ affecting kind of new grad job placements. And so I think my intuition is it’s gonna spill over to kind of the mid-level folks.as well, right, comparatively soon.Grace Shao (44:26)Yeah, I think actually I have an episode coming out literally today, which is with ⁓ Diana David. She’s the director of features at ServiceNow. And she was saying kind of like, instead of thinking about how it’s going to displace jobs, it’s going to change where the workforce will go towards. It’s just that we need that time to kind of figure out where it’s going. But like you said, right now, unfortunately, it’s hitting the young junior staff the most because their work is usually, you know, moreyou know, the hands-on kind of like the grunt work that is really easily done right now by AI. Anyway, thank you so much for your time today. I have one last question for you, which is a question I ask everyone. What is a view you hold that is unconventional or you think it’s against consensus? It could be something related to investing or anything, you know, in life.Kevin Zhang (45:49)Let me try to think something that’s unconventional.Okay, ⁓ so I think something that’s unconventional, especially folks ⁓ within finance, think folks in finance tend to be very sensitive to kind of market fluctuations as well some of the macro headwinds or tailwinds. And so, especially right now, where I think we’re in potentially a more challenging part of the cycle.with employment, with kind of the degradation of ⁓ kind of global connection. think in the long run, ⁓ collaboration will probably win out, at least that’s my hope. ⁓ And that bodes well for entrepreneurs, right, who want to play in multiple ecosystems. That bodes well for investors who want to play at ⁓these multiple ecosystems right across across Asia Europe and in North America and I think ⁓ the game is Right now is especially for for some of these venture firms in China is like who can survive right because I think those that can survive and Those who can continue to raise they’re gonna be fine and they’re gonna they’re gonna see some very very good vintages ⁓ and I think the same goes for ⁓when the US also sees their down cycle, the folks that have a lot of dry powder who are able to deploy through those tougher periods, think that’s where you’ll see good ventures as well. so the question of like, given some of these macro headwinds, is that the death of the buy side in Asia, I think that’s probably a little bit overblown.And then like I five years from now, 10 years from now, like the ecosystem will be mature and healthy.Grace Shao (47:53)You think that USD denominated funds will see a revitalization or do you think that we’ll just see a complete different kind of ecosystem from five, 10 years ago?Kevin Zhang (47:59)Maybe?I think it either might be a completely different ecosystem or ⁓ there’s other global capital that’s interested in the broader Asia ecosystem. And I think they’ll come. It may not be at the same scale as ⁓ the era of like Hulianwang, right? But like as long as the ecosystem continues to be vibrant, as long as ⁓ entrepreneurs are able to buildgood products and generate or build large businesses, then there will be some capital somewhere in the world that’s willing to back these entrepreneurs.Grace Shao (48:46)There’s a lot of Middle Eastern money coming into China, actually, and I think Southeast Asia, family office money, and even, I think, European money. But it’s definitely not the same scale as what we saw during the Indian era with institutional US investors, right? But that’s encouraging. That’s positive note to end on.Kevin Zhang (48:52)Thank you.Yeah.Yeah, yeah. I think like for some of these LPs, right, there’s definitely a learning curve, right, in terms of LP sophistication and being very long-term oriented. Whereas ⁓ some of these ⁓ OG US LPs, they’ve had decades of experience investing in the VC asset class, right? And so when that learning curve catches up, then... ⁓At least on the LP to GP side, ⁓ we’ll see some revival.Grace Shao (49:39)Great, thank you so much, Kevin. Kevin Jang, thank you.Kevin Zhang (49:42)Great, thank you.AI Proem is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to AI Proem at aiproem.substack.com/subscribe

In this conversation, Jing Yang, Asia Bureau Chief of The Information, a former WSJ reporter, discusses the evolution of China's tech landscape over the past decade.We explore the corporate strategy and positioning differences between established tech giants like Baidu, Alibaba, and Tencent and newer entrants like Pinduoduo and Shein. Jing also talks about her reporting on Shein and Temu and their attempts to be publicly listed in the West.The conversation delves into the regulatory challenges faced by these companies both domestically and internationally, and how that has led to a belief shared amongst Western investors that “China is uninvestible.”She dives into the implications of the AI arms race between China and the US, and the shifting dynamics in the venture capital landscape in China. She explains the differences between RMB-denominated funds and US-dollar-denominated funds, as well as how the VC ecosystem has evolved over the last few years.—Jing Yang is the Asia Bureau Chief at The Information. Based in Hong Kong, Jing leads a team of reporters covering the region's vibrant tech and venture capital scene and has written and overseen agenda-setting stories on topics ranging from AI to semiconductors to marquee companies like Nvidia and ByteDance. Prior to joining The Information, Jing was a Senior Correspondent at The Wall Street Journal where she covered a broad range of topics, including Wall Street’ foray into China, Beijing’s crackdown on internet platforms, and the 2022 Beijing Winter Olympics. Jing also has reporting stints at Bloomberg News and the South China Morning Post. She has won three Society of Publishers in Asia Awards and three Best in Business Awards at the Society for Advancing Business Editing and Writing in the US. She is an honorary lecturer at her alma mater, the University of Hong Kong’s Journalism and Media Studies Centre, and a board member of the Foreign Correspondents’ Club in Hong Kong.In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.Chapters00:00 Introduction to Jing Yang and Her Journey02:35 The Evolution of Chinese Tech Mindset08:08 Comparing Old and New Chinese Tech Giants11:16 The Unique Business Models of Pinduoduo and Shein17:41 Regulatory Challenges for Chinese Tech Companies22:04 The Impact of AI on Chinese Tech Giants32:30 US-China AI Arms Race: Context and Implications34:55 Bridging the Gap: US and China Perspectives39:10 China's AI Strategy: Open Source vs Closed Source44:32 Emerging Players in China's AI Landscape51:41 The Impact of Decoupling on VC Landscape: Sequioa to HongShan01:05:44 Future Trends in China's AI Tech SpaceAuto generated transcript for reference, typos might occurGrace Shao (00:00)Hi, this is your host Grace Shao and joining me today is Jing Yang, Asia Bureau Chief of Information. Previously, she was a senior correspondent at the Wall Street Journal in Hong Kong where she covered a range of topics including Wall Street’s foray into China, the globalization of China’s most prominent tech companies, and the country’s domestic tech regulatory crackdown. Jing, it’s so great to have you on and ⁓ it was lovely running into you actually recently at Karen’s book launch in Hong Kong. We’ve known each other for a few years now, you know, over like our paths crossing.whether in journalism or when I was working in Alibaba, you’ve covered Chinese business and tech for over a decade across from SCMP, Bloomberg, Wall Street Journal, and now obviously leading the information out here in Asia. So looking back right now, like over the last decade, what’s really changed the most in terms of talent, capital, and company building China, and just your own kind of observations of the whole industry?Jing Yang (00:52)Sure. Thank you for having me. Yeah, it was indeed good to run into you at Karen (Hao)’s book talk. So I think the one thing that sort of really struck me is that when I started covering the, started ⁓ working as a reporter, 10 plus years ago, at that time, if you recall, then China was rapidly still integrating into sort of the West led global order.A lot of the Chinese companies and executives and entrepreneurs, think their mindset was still like there’s so much we can learn from the West, from the US, from Europe, from what companies have achieved there. And then now that has changed for various reasons we can get into later into sort of, know, this is the way actually, you know, we do things in China and it has in so many cases proved to be actually working better oror there’s more this awareness of things are different in China, whether it’s cultural or economical or social. And I think then you also have seen sort of equally some examples from Silicon Valley or elsewhere sort of looking eastward and looking at what are the things that China or Chinese companies have done that we may actuallythere are some lessons or experiences we can draw from. I think that is a very interesting sort of shift. It’s not just one way anymore, it’s more two-way.Grace Shao (02:17)For sure. But I think in that sense still, you know, we have people like Bill Gurley going on his podcast saying like Chinese entrepreneurs or business leaders are just so much more attuned to what’s happening in the U.S. compared to vice versa. But there’s definitely more more interest coming from the West to China. Right. I think on that, like one interesting, you know, Chinese tech culture that’s really being adopted and, you know, seeing it being embraced by Silicon Valley right now is the 996. You know, likeThat’s an interesting phenomenon. What do you think of that? Do think people are realizing that you got a 996 to really push ahead?Jing Yang (02:50)Yeah, that is a very good question. And I think I’d like to actually unpack what we are talking about when we’re talking about 996. When it first started in China, and by the way, that wasn’t just something adopted by tech companies, but also in general, a lot of big companies in the private sector in China were doing that. And I think what in that sense, from the mid to late 2010s,When companies were adopting 996, it was mostly codified. Remember, that was a pre-COVID world. You are expected, when you’re working, you’re expected to show up in the office, at your booth or whatever. So that means you have to clock in at 9 a.m., clock out at 9 p.m., and then for six days a week. But why you actually do in those 12 hours and six days a week is another question.And I think over time, some companies realize there’s actually some kind of waste and inefficiency that this codified system has created. For example, a lot of companies, ⁓ employers, in order to entice employees to work longer hours, they provide dinner and then they provide, say, you can expense your cabaret home if you work past 9 p.m., that kind of thing. And over time, then, I think that actually did lead to some sort of inefficiency in the sense thatMaybe some people just want to stay back. If I’m going to go home and have a take on myself, then why don’t I do it at the office when it’s paid for? I’m not saying everybody does that, right? But certainly there are people who did that. And I think while I’m not discounting at all, Codefine 996 has led to burnout and even much worse things. But on the flip side.is that when you codify something like that, that is the problem. So then if we look at in the US, for example, if you are a Wall Street banker or lawyer or a management consultant, I think you are expected to work long hours when you are servicing a client who has urgent needs or you are just on some really tight project deadline or say if there is a market meltdown.And then nobody would say anything about, that is 996 or not. And then so I think it’s easy to throw the labels around and then not to actually look at what is happening. It’s not like in the US or anywhere in this world that people just don’t work extra hours. That’s certainly not true. And in terms of, I read the reports. I think there was a wide article that wrote in depth about this.But the one thing that I think is quite different is that we haven’t seen this being adopted as a policy across-broad, especially by big companies in the US or anywhere else. And in China, when they loosened the 996, what was loosened was just like you are not expected to clock in and out at 9 a.m. and 9 p.m. But you do still expect to work when needed.And then, and I think in a way, the post COVID world where most workplaces is set up to have employees work, you know, virtually from anywhere has actually made it worse for people everywhere, right? You know, as long as you still have your laptop or even your phone with you, you’re expected to say if your boss messages you at, you know, 10 PM with some questions.I think it’ll be unthinkable that I just ignore that, right? And so I think that is actually the real issue than when we say, you know, 9-8-6.Grace Shao (06:26)Yeah, think it like I totally agree with you in many ways that like this whole working from home thing has actually made everyone feel like they’re more like glued to their devices and more clocked in. But I think it also just showcases that this whole adoption of so-called 996 in the US right now is really driven by, you know, just the excitement and I guess competition now we’re seeing in AI because, you know, for big tech for many years, we’re used to this kind of story of like, you know, ⁓not a very like, it’s a very cushy job, not that competitive internally once you get in, you get all the perks of like the gyms and the free food and everything and people can just kind of go for walks in the middle of afternoon, right? and that culture’s completely changed. And I do want to really double click on that on the China US AI arms race per se later on. But before we get into that, I really want to kind of look backwards first, try to start of your career, right?You’ve been covering China’s tech and business space for more than 10 years, essentially. Comparing the earlier BATs, whether it’s the Baidu, Baidans, Alibaba, Tencent ones, to the new companies like the Pinduoduo, Shianti, Moos, what are the kind of differences in their corporate structures or management styles or even their so-called Chu Hai strategy, like their going global strategy? What are some observations on that front?Jing Yang (07:44)Yeah, I think overall, I would say, you know, since you mentioned BAT, I think what the B-stands for Baidu are Bytedance, right? But if you look at just the A&T, you know, the Jack Ma and the Pony Ma, I think they represent sort of the last generation of Chinese entrepreneurship. And then moving on to like Jiang Yimin, who’s younger, and then obviously, Kuanlin Huang as well. And then to like the...the founders of the hottest startups nowadays in China. think one big shift is that the entrepreneurs, the builders who grew up, who were born post 1985, post 1990, they grew up in an era when China was rapidly integrating with the world.And they likely grew up watching a lot of American TV, really engrossing a lot of American culture. then that sort of... And it was also during that time when China was really just thinking like, we are the students learning from the superpowers from the West, that kind of era, right? So they sort of grew up to have a more global view.of the last generation of founders and entrepreneurs. then so now this is why the companies that they are trying to build right from Bydowns, which is already a giant today, to the smaller startups that we’re seeing, they want to compete at a global level from day one. Whereas in the case of Tencent, Alibaba, or Baidu, what they did was wetook something that has worked on, for example, Amazon or Paypal in the US, and then we replicate that model, adapting it to the Chinese soil. And then we make it work first in China. Then if that has worked, then let’s see if we can export that success somewhere else. I think that that’s probably the biggest difference.Grace Shao (09:41)And I know that beyond the BAT, you know, covered actually, Temu and SHEIN quite extensively. I feel like these two companies are a bit more mystical to the world because, know, they feel like in some ways they were the first companies that really succeeded at not being seen as Chinese, even when they went global. Yet they were caught in the geopolitical rows when the first round of trade wars happened. And, you know, now they’re not really making the front of the headlines anymore. So what do you really make of these two companies?I guess their positioning, how they’re doing now and their journey to potentially trying to get listed in the West and just frankly failing at it at this point or not being able to.Jing Yang (10:21)Yeah, I mean, maybe we’ll just separate PDD slash Temu and SHEIN a little bit because as you know, Temu is a subsidiary of PDD, right? So I think Xin is a very interesting company in the sense that it’s kind of, I mean, of course, you know, people see them as an e-commerce company. But however, if you look back at the roots of the company and the background of their founders, it’s not really a tech company in the sense that, you know, they arevery good at doing international trade, foreign trade in China. And then e-commerce is just internet, it’s just a way through which they made their, it was like a marketing and a sales channel for them. But the company didn’t start as like say with tech or internet in its genes. It was...For example, in the early years of SHEIN, they really basically went to like all these markets in Guangzhou or somewhere else and then see what we can sell to overseas markets. And then ⁓ they’ve sold a bunch of random things from teapots to wedding gowns and then, know, basically whatever.sold, they will sell it. ⁓Grace Shao (11:33)It was like a glorifieddrop shipping business essentially, right? Like a bigger scale drop shipping business essentially in the beginning.Jing Yang (11:39)Yeah, exactly.So that’s how the company got started. But obviously, they then fine-tuned their business model over the years and then did make some true innovations in terms of how to adopt technology and making fast fashion more, not only more efficient, but also more trendy and also through the process, it really drove down the cost.But all of that was happening against the backdrop of China really has achieved that kind of manufacturing base that enabled this kind of opportunity to arise. And I think that is the part of the Shenzhen’s business model that was most misunderstood and why there was all these allegations of slavery for slavery. Because it’s just the...the cost efficiency was just mind blowing to a lot of people. And then obviously what Temu has done is to expand, know, Shin’s business model, replicate that from not just apparel, but to many other sectors. But essentially these two companies and or as e-commerce platforms per se, they’re actually quiteyou know, different in a sense, because what SHEIN has been selling, apparel, is like a pretty non-standardized kind of merchandise, right? Every piece of clothing is different. And then they change so rapidly in terms of what is the market trend. Whereas when you are selling, say, toys or home appliances, these are standard goods that don’t really change that much over time. And also,⁓ the marketing and the shipping and everything is also a lot, is quite different. But in terms of like sort of how they were misunderstood, I think the rise of SHEIN team was really to a large extent misunderstood in the West because what I said earlier about, know, because those kind of business models was enabled because of the decades of, you know, China being the worst factory reallyaccumulated the manufacturing capabilities. when I say manufacturing capabilities, doesn’t just mean factory, the efficiency of the factory floors, but also everything that goes behind it, you know, the infrastructure, the logistics, etc. And then in the process of that rise, I think in the typical Chinese mindset is like, I don’t want to talk about my success or how great I am. I just want toYou know, this very, maybe very pure mindset. I just want to make money. I just want to build a successful business. Right. I only need to satisfy my customers. Like my customers are the only stakeholder I need to think about. Right. They don’t think about, there is also, you know, investors or potential investors, regulators, politicians.I used to joke back when Shane started to face all this backlash in the US. I used to joke that the people who came out against criticizing Shane in the US are like the pirates of their customers. That shows you sort of where the mismatch is.Grace Shao (14:52)Yeah, I think it’s so interesting that you brought that stakeholder engagement part up and it’s really funny, I think. I remember SHEIN when they were like, they made it to the front page of Bloomberg saying, like you said, being accused for labor malpractice or whatnot. They were just trying to hire so like crazy, like frantically hire people who can manage their PR. But it was kind of one of those things where frankly it was a bit too late. Like they really didn’t have the sense to.actually put out their messaging and put out that what like explain what they’re doing beforehand. But I actually do want to kind of bring it back to the regulatory side of things. So basically, where are we at right now with the two companies? ⁓ Like, well, I guess, and more on the team side, as far as our she inside, are they still trying to pursue a IPO in the West? Or, you know, what where we are right now? And what’s really the hurdle? Is it like an international?kind of a geopolitical hurdle, or do you think it’s actually a domestic regulatory?Jing Yang (15:50)Right. I think it’s sort of a both. I mean, I’m just like repeating what has been reported out there that SHEIN now as a confidentially submitted application to listen in Hong Kong. By the way, that is sort of an exception made by the Hong Kong Stock Exchange because usually, like typically the only times that the Hong Kong listing regulatory regime does not usually allow confidential filing unlike in the US and in the times that they wouldusually ground that exemption is for companies that already listed elsewhere. You remember all that homecoming listing wave that happened a few years ago, Alibaba and Baidu, cetera, that when they were applying to list in Hong Kong, they were all exempted to file confidentially. And the Hsing, in that sense, being a company that is not listed elsewhere, that is a of a waiver that the Hong Kong Stock Exchange gave them. And then in terms of hurdles, mean,All of this happened after the Didi debacle, right? And after which CSRC tightened the regulatory framework for companies, for Chinese companies taking the list overseas. And then SHEIN sort of was caught in, know, SHEIN and many other companies IPO have suffered significant delays because of that.And in SHEIN’s case, because the company is really quite big in terms of the size and also all the attention it has attracted. But if you look at from, like, say, a pure domestic Chinese angle, what is what SHEIN is as a company? It really, as I said earlier, in the China domestic angle, SHEIN is an employer. SHEIN is a company that that is a big customer to a lot of factories in China.That’s essentially what it is, because it doesn’t sell in China. So then in that sense, SHEIN is a very big contributor in terms of the whole economy that it has given rise to, as well as the tax dollar, the number of employment that either directly and indirectly has contributed. And then that sort of makes it like a case.Grace Shao (17:34)Exactly.Jing Yang (17:57)in the regulatory context, that case that cannot be, that has to be really, say, deliberated on. But then does China want a company like this to live, in the US or in London? I think that’s why there is the interplay between the domestic consideration and the geopolitical tensions and also the backlash to that.that came with it.Grace Shao (18:20)Yeah, I think that explains it really well. think for a lot of people outside of China, they don’t realize SHEIN actually is not like a household name at all. The people, the consumers actually are, it’s not a consumer facing company in that sense in China at all. It’s actually to be business per se in China, right? I think it’s perfect. Exactly. I think it’s great you brought up the Didi (Chuxing) situation and I think that’s where I want to kind of bring it back to last question on the big tech space in China, which is like, you know,Jing Yang (18:32)Exactly.is like a B2B wholesale company, essentially.Grace Shao (18:47)We all know there was a domestic, domestic regulatory kind of tightening over, you know, between 2020, 2023 per se. know, Didi being kind of at the very top of the epitome of what was happening. And then Baba, Tencent being faced with the, situation as well, right? Like the two choose one between the two kind of camp situation, the regulatory problems. So, but after that.Basically international investors pulled out of China. People were kind of scared of the China regulatory crackdown. think people outside of China don’t really fully understand why the domestic regulators were cracking down on these companies. Could you give some context on that? And then I think what I really want to also ask you, the second part of the question is, are we seeing a revival of these companies now withAI being supercharged into their strategies? know, recent BABA and tens of earnings have done really well, all driven by AI, right? Their new AI strategy. Or are they kind of just, you know, the last generation staying there back in 2023, they just kind of stalled and stopped there? Are they becoming relevant again? So I think it’s two parts of the question.Jing Yang (19:52)Yeah, so yeah, let me tackle the first part. I try my best even back when I started, you know, this whole thing happened when I was still at the Wall Street Journal. And back then, I think in the beginning, it’s been called the China tech crackdown a lot. And I actually made a point when I was writing about it at the journal, after a while, it has become very clear it’s not a tech crackdown. It’s not acrackdown on tech companies. Because if you look at sort of the hard tech companies, either it’s a Huawei or any other that’s in that space, they were all fine, right? Essentially, the crackdown was targeted at internet platforms, right? Which sort of is equivalent of big tech in Silicon Valley or in the US when we talk about it. But in China, there is actually a differentiation betweenInternet platforms and tech companies. So that’s the first thing. And secondly, I think in terms of that regulatory assault, there are, you know, there are, you know, legitimate sort of economical and regulatory reasons to do so. Right. As you mentioned, the Arsheng, the truth one from two, that was that was indeed a violation of China’s antitrust regulation wheree-commerce merchants were forced to only choose to sell their wares between Alibaba and JD, for example. when this whole thing happened, was indeed sort of the regulatory incentive to do that. was indeed raining years of flouting.anti-trust regulation and other types of regulation. But then obviously, the problem that came with it is that the way that the Chinese regulatory framework and the Chinese government in general works is that it doesn’t really care about doing a very good job at telegraphing its regulatory intention.That’s not just on the tech space, like overall, So that’s why if you remember the online tutoring crackdown, Just essentially what happened was just, that was very scarring, by the way, for a lot of investors, right? Because they targeted a sector where the biggest companies are listed in the US.So then when you can’t just do this, issuing a piece of document that just evaporated the entire sector, that decimated the entire sector overnight. That was very scary. ⁓Grace Shao (22:31)Yeah, it was so sudden. That’s what it was. There was no hints or clues. I feel like with the anti-monopoly, was actually, you know, there was like momentum building up to it.Jing Yang (22:41)Yeah, so that really, I think in my recollection, that really crystallized the saying that China is uninvestable. When people say China is uninvestable, they mean Chinese stocks are uninvestable because that really crystallized how precarious, so to speak, that Chinese policymaking can be. And then so...Similar thing with the Didi debacle. think the biggest problem with that is, one can argue on Didi’s behalf that the company didn’t do anything wrong because what happened at the time was there really was just no regulation governing Chinese companies listing overseas. Things were just not formulated. And obviously, Chinese regulators realized that was the problem. that’s why they...came up with this new framework that has been in effect since early 2023. However, the absence of a formulated framework did not stop them from punishing companies in the first place for flouting rules that are not explicit. Then that is another piece that shows, OK, so this is unpredictable.this whole virtual environment.Grace Shao (23:58)And the second part of the question is, you think we’re seeing a comeback? Because essentially, now you can’t really get into China’s private AI sector, right? If you’re a foreign investor and the way they can get some kind of exposure, I guess, is through the US listed companies, tech companies. In this case, it would be the Alibaba’s of the world, right? So are we seeing kind of a shift in interest again, or a revival of these tech giants?Jing Yang (24:23)Yeah, I like to answer that question by going a little bit further back first. So you remember the Shanghai Stock Exchange when they first came out with the NASDAQ-style starboard, right? There was a lot of discussions on what are qualified as innovation so that a company can qualify as being listed on starboard.Grace Shao (24:36)Mm-mm.Jing Yang (24:47)And you remember Xiaomi actually was going to become the first company and then it didn’t happen. And I remember having conversations because I was covering IPO and capital markets of Wall Street Journal at the time. I remember having conversations with some lawyers who consulted with who advised the Shanghai Stock Exchange on designing the rules for Starboard. And there were still a lot of undecided issues such as, OK, for example, did the other time remain unlisted?Then the conversation was like, Didi as a company, even though it wasn’t an innovation in terms of the technology of like, ride-hailing, for example, but it was innovation in the business model. Then does innovation in business model qualify as true innovation, therefore qualified being listed on Starboard? There was all these questions at the time. And obviously, we now know the answer with what happened with Didi later on.And I’m bringing this up because shortly after the crackdown was on the Internet platforms, followed ⁓ the zero COVID and the whole Chinese economy and many other things related to that just were in a really depressing, know, slipped into a really depressing state. ThenAround that time, toward the end of zero COVID, had the arrival of Chagabitie that sort of shocked the core of a lot of tech companies and researchers and investors in China, which we can get into later. But I think what the crackdown on internet platforms made people believe that actually China’s paramount leader, Xi Jinping, is probably not a fan of internet platforms.He probably does not think the type of innovation in business model that I just talked about qualifies as true innovation. Instead, the things that achieve their companies at Huawei are true innovation, are the real technological advancement that can help bring China forward. And so I think that coupled with the arrival of ChatGPT,sort of, you know, really served as a wake up call, I think, to both the tech incumbents and the startups in China that we really need to, you know, we are actually behind. We have been, you know, because you remember like in the mid 2010s, you know, China had this four AI dragons in the computer vision age. And then you have a lot of people proclaiming that China is ahead of the US now in terms of AI, right? And obviously, that was all.That sort of dream or awareness was just shattered. All things came together between late 2022 and early 2023. And then that’s where we are now. So it’s hard to tell whether when we see the BATs nowadays...really doing some of them doing really good work and innovation in AI is driven by the regulatory reason or else. But I would say just the way that things have played out seem to point out to the direction that this really is the era that they have to go through.Grace Shao (28:01)Yeah, like to your point, I think there was like an awakening where the focus or the wanted focus was on the so-called hard hard power competition, right? And like you said, that the kind of slippery slope downward trajectory for the industry was really not just caused by one thing, but it was just the macro situation, the regulatory situation, the companies also not innovating in some people’s eyes, were not doing tech for good for the community somewhat.You remember the common prosperity rolled down, right? All these things are kind of just adding up together. And then there was the whole COVID zero thing that just really made everything even worse. think that that definitely fed into the fear for international investors and international, I guess, China watchers per se, if you have to put it that way. But I think I want to really now go into the next section of our conversation, which is what you’ve kind of touched on already, which is China and AI, right? And China.Jing Yang (28:31)yeah.Grace Shao (28:57)China’s AI space and how China is positioning itself right now. I think from the West, especially Western media, it’s very, very, very much focused on this idea of arms race between China and the US. You know, there’s a lot of ⁓ comments about how deep the deep sea moment woke Silicon Valley up, made people realize that China was catching up, you know, there’s some fear mongering, frankly, I think, but there’s someI think some charged by actual fear or confusion or even surprise. How do you kind of make up that? Like, I guess just a high level context first before we go into details.Jing Yang (29:34)Yeah, mean, truly, when we talk about US-China AI race, we cannot talk about without talking about US-China competition, whether it’s, I think people now generally call them this strategic rivalry. That is all happening against this broad backdrop. And then the one thing I would sort of⁓ note though is that, you know, it’s not like there’s a lot of companies or builders or funders in China from day one thinking about, I want to like outcompete ChatGPT or, you know, Anthropic with what I’m doing right now. I don’t think that’s... Yeah.Grace Shao (30:13)Yeah, that’s the point. Yeah, like, like, there’s not that strong narrative in China domestically, right? And I thinkthe information you guys because you guys write for such a frankly, sophisticated audience and people who are kind of more focused on really the business. I feel like your coverage is not as geopolitically charged. It’s actually just talking about what the innovation is, how the capital market is moving. So I can think from your perspective, like, what how do we make a business? this all noise? Is it just likebecause of American domestic political reasons that there’s this kind of geopolitical narrative? Or do you think there’s some sense that in the tech space in the US, they genuinely see China as a rivalry that they have to hold down versus like we said, like the China tech space actually just, they don’t talk about that as much. If anything, I think the China tech space actually admires American tech space quite a lot, like as in they really worship a lot of business leaders. They study their business models, you know, like there’s less of a rivalry sense, right?Jing Yang (31:08)Yeah, so I think in the US there is a bit of a boast of what you just talked about. I’m not an expert on America, but what I have observed, at least the China relevant people that I talked to from Silicon Valley to Washington DC, I do sort of see like increasing bridging. I think in the past, you know,Silicon Valley, obviously, we know is more left leaning and then, know, this is different. then with, you know, Trump’s presidency, you see all of that, you know, that that gap is bridging. And that actually ended. And when I say the gap bridging, the gap is also bridging when it comes to the China discourse. So on one hand, there is definitely fear mongering. And that fear mongering, think previously probably mostly concentrated in the D.C. and now is like spreading to Silicon Valley. That’s what I.observed, right? And also, when I was talking about in China, how people were shattered with the release of ChatGPT in late 2022. Equally, think in the US, what do we see about 12 months later is that with the advent of a DeepSeek R1, a lot of people in the US are like, know, America is far ahead, is the indisputable global leader in AI.from the China-GDP moment to then the deep sea moment is that China is faster catching up or in some cases they even say China is already ahead. And also that in particular is very true, the fear mongering in the semiconductor space as well, because we can’t talk about it without talking about semi. So I’ll just be very quickly talking about here.from Jensen Huang to like many other people. I think the last year or so they have been talking up of Huawei and other homegrown semiconductor companies in China and the capabilities of their chips. But the reality, I’m not saying Huawei is not progressing, but the reality as our reporting has shown.the information. The reality is that actually the BATs themselves actually do not want to adopt a Huawei’s ascent chips for various reasons. First is just the tech is just still not there. Nvidia is still the gold standard. And the second is also because all these companies compete with Huawei in the cloud computing business as well. Like why do you want to enable a big competitor? But you don’t see this being talked about when in the US.From Silicon Valley to DC, when people talk about how Huawei is threatening Nvidia, how US semi control policy has enabled, has forced China to compete and innovate faster. So I hope that makes sense, by the way.Grace Shao (33:52)Yeah, yeah, definitely. think that’s something I wrote about as well. just like, it’s actually people are not realizing from a very selfish business perspective. At the of the day, these companies are for profit. they’re public listed companies. Their goal is to make money. They’re not like, you know, following government orders to like, know, create something on a national level for the sake of that. So like, to your point, like they don’t want to give money and give business to Huawei because essentially one of biggest competitors, right?Jing Yang (34:15)Yeah.Grace Shao (34:20)because like Baba and Tencent, they all have their own cloud business. I think I want to double click on one thing. said, know, there was, when Deepseek came out, R1 came out, well, V3 and then R1 back to back, know, Silicon Valley said, maybe China is ahead in some ways, right? At that point, it was talking about the engineering and efficiency, You know, from your reporting, where is China actually ahead?behind or really differentiated in terms of their whole AI strategy? And this is a broad question. So it could be about the companies particularly, or do you think the whole ecosystem is operating a different way? How do you see that?Jing Yang (34:55)I think obviously DeepSeek sort of made it cool, made it the open source and open weight school, sort of school of thinking. Cool, right? I think if you remember before DeepSeek became phenomenal, from Baidu to Alibaba to Bydance, everybody was just doing their own closed-source models.If you do open source, only release smaller or more inferior models. You only open source those. And then now that what happened with DeepSeek, just really made in the LLM space, made people realize actually China can be ahead. China can have a real edge if it pushes ahead with open source. And thenAnd then that and also when you open source your model models, it also just naturally encourages like a broader and wider adoption of your models. It sort of can travel beyond the national borders on its own. Right. And if China were to let’s say if, you know, by the way, as we established, right, it’s not like the Chinese companies that are building large language models are thinking about what the policy.or what the government officials are telling us. They really truly are just thinking about, like, I want to be better, right? How do I get... If you build a model, obviously you want your model to be used. You want developers to build applications on top of your models, right? You want the cloud providers to include your models as part of their offering. So I think that’s what’s driven that. And I think the open source thing really sort of shown that, you DeepSeek made it...a lot people realize that a lot of people in China realize actually open source is the way for China to pull ahead. And that has happened obviously with Alibaba’s Qwen and also I think most recently, Zhipu as well as Moonshot came up with their own latest models that also have impressed a lot of AI watchers in the US as well. So I think that’s where,That is where China is different. And I just want to add one more point as well. In the US, there’s a lot of money to be made in providing enterprise software solutions. then naturally, that’s when AI applications are being built. You want to build for survival. You want to build to be able to become profitable.naturally the enterprise software solutions, right? You build a 2B business. But in China, it’s completely the opposite. Chinese companies in general just are very stingy in terms of paying for software, right? And then so that sort of is the way the ecosystem works. And it has always been like that, right?enterprise customers just naturally go to the next cheaper solution. then, you know, entrepreneurs and also venture capitalists that see who see these companies know that this is the environment, then then they also know that, you know, if you go down that route, you’re just basically, you know, waiting to to go out of business. And then so that’s also what makes the open source led, you know,an open source led business model, you know, like have a better chances of working in China. Right.Grace Shao (38:04)Yeah, definitely. I think the open source versus closed source discussion has been even, you know, rooted out in the software era. like, to your point in China, it’s really like a market share like business. you it’s like you commoditize open source models, you try to capture all the market share, right? Like it’s it’s very consumer facing right now. A lot of the LLM they’re all rolling on consumer applications versus the US to your point. You know, it’s it’s you actually make you try to make higher margins on a lot of these enterprise products, but you just can’t in China.like the willingness to pay is still so low and not just trying to buy things across Asia in general, like the willingness to pay or I don’t know if it’s a cultural thing or it’s just like people just don’t want to pay. you know, I was even shocked I think when I was working for one of the big companies and like a Chinese company for a while where you would even have like pirated software in their company computers. like, ⁓ you definitely don’t need to save from that money. But it’s just a mentality and then there’s a lack of like.I must pay for privacy, whatever mentality around it. So culturally, that definitely has shaped the markets quite differently. So I think we touched on quite a few of the startups. So like you mentioned, was Drupal, Bytron, Moonshot, Minimax. Do you think there are any companies that are kind of going unnoticed still in the West? Like we named the four that are still called, there’s still a...call it what was the four dragons at one point? Wait, four tigers. I get the mixed up with just tigers, dragons.Jing Yang (39:30)Six little dragons. Yeah. Yeah.Grace Shao (39:33)Yeah, they keep on updating them. It’slike new versions of Dragons and Tigers. are there any online companies right now that you think are going kind of under a notice? Because DeepSeek in many ways actually was not being, they don’t have strong PR. Liang Wenfeng clearly is a very low profile guy. And I don’t think people really knew about him outside of China and China’s AI ecosystem until V3 came out. So like, are there any kind of...Jing Yang (39:38)Yay.Grace Shao (39:59)companies that you’re eyeing or covering that you think could be the next one, or you think the LLM space is already pretty saturated and incumbents are going to kind of stay as leaders at this point, or we might use the consolidation.Jing Yang (40:10)Yeah, so back when we talking about the six little dragons or tigers or whatever, DeepSeek was part of it, of the six, but it was probably the least talked about because it’s just very different from all the other five, right? All the other five had half venture capital funding, have outside investors. DeepSeek remains fully funded by High Flyer Capital Management till this day. But if we’re limiting the...the scope to just the LLM developers and I would say that I don’t see the possibility of having anyone that we haven’t seen out there. As a matter of fact, I’m actually surprised that the consolidation hasn’t happened at a bigger scale with the exception of of Alibaba essentially acquiring 0.1.AI. We haven’t simply seen otherconsolidation happened yet. And I’m actually surprised. If you ask.Grace Shao (41:05)For context,sorry, readers, sorry, listeners, that one’s the one that Lee Kaifu founded. outside of Tsinghua campus, yes.Jing Yang (41:09)Yes.Yes. And then now we see that Jhipu has filed to or is sort of looking at to go IPO in both, you know, Chinese, Asia market and Hong Kong. You know, I think whenever they release their prospectus, people will be reading it with a lot of interest. Other than that, you know, I’m just surprised that the so-called dachang, the tech incumbents in China have not consolidated.I think there was a time in my, you know, I recall in my reporting, like say in my conversations with sources about 12 months ago, there was a time when some tech companies were seriously thinking of acquiring some of the companies which is mentioned. And for one reason or the other that didn’t happen. And all the tech companies in China actually decided to build their own as well, right? So this is quite different from Silicon Valley if you look at it, right?we’ve seen from Google to the other companies, are crying pretty sizable startups, or not really are crying, are quite higher. So I do think that the Chinese LLM space doesn’t need some consolidation. It’s just not sustainable. need the level of compute that is needed and coupled with the chipshortage China finds itself in. It’s just not sustainable.Grace Shao (42:32)It’s such a high capex game here, so how do these startups continue to fund themselves? I want to go into the VC space later, but mean, end of the day, it’s only the big check comes to have the money to keep on even chucking into that machine.Jing Yang (42:39)Yes.Yeah,that’s exactly what I was trying to get to, just not sustainable to have, say, 10, 12 LLM developers.Grace Shao (42:55)Exactly. I think it’sinteresting, a lot of them are financially backed by them, but not at a very big scale at this point. It’s like, you know, like a hundred million dollars here and there, but nothing bigger than that rate.Jing Yang (43:08)Yeah, exactly. if I were to, I Alibaba has probably backed it the most, right? And obviously, it did it also for the sake of promoting, know, helping the cloud business, you know, expand its market share as well. But if I were to take a guess, I would say that some of these companies will just have to really run out of money.And then to the point that existing investors are willing to sell, say, don’t know, five cents on the dollar or whatever. So that it becomes financially attractive for any of the tech companies to actually acquire them. Otherwise, you know, it’s just not going to, you know, I don’t know the tech companies. They’re all very, you know, the people sitting on top of this company are all very like sophisticated. They’re not just going to be, you know, spending like, I don’t know.20 billion dollars or even not 20 billion, like 10 billion dollars or five billion dollars by a technology that they think maybe they already have on their own, or they can do better on their own. So the only way that it can work is that, you know, things have to just drag along a little bit longer for the pricing and expectation to match. Right now, I think there’s a pretty big gap. And also, lot of the entrepreneurs of theGrace Shao (44:21)Yeah.Jing Yang (44:25)among the six little dragon camp. But they also don’t want to call it quits yet, I think.Grace Shao (44:30)Yeah, I think that people are holding on to their dream. And I think I like people ask me about what I think of the AI startup founders of this generation as well. Like to your point earlier, they kind of grew, they grew up differently from the last generation entrepreneurs. And there’s less of a, want to make a quick buck kind of mentality. They are really much more mission driven when you, you know, like hear, hear about them or, know, their media appearances or what they say out loud, you know, like where you’ve been the moonshot guy, Yang Zhilin really, really just talking about how much of aknow, philosophical pursuit it is for him to chase AGI. Actually, you know, I want to kind of pivot into the VC space. We touched on the fact that China’s VC space, frankly, is not as vibrant right now. You know, obviously the majority of the money right now, even to back these startups are from the VATs or the situation like any of the bigger internet companies from the last generation. So how has the regulatory resetsince 2021 really reshaped the local domestic VC landscape because we saw that Sequoia was probably one of first that started the decoupling effort, splitting out their Chinese business called Hongshan, which is like a literal translation to Sequoia, the tree. Are we continuing to see this play out or do you think we’ve kind of finished it, wrap that up already? It’s been a couple of years and the VC space is getting a bit of an energy back or...recapitalization.Jing Yang (45:53)Yeah, I’ll talk about decoupling first. I I think the US China venture capital decoupling is already over, right? Like it’s already decoupled. And it’s decoupled mostly for driven by factors out of the US, out of American politics, right? When the Treasury Department in Biden’s final month, in the Biden administration’s final month,came up with this long expected rule that essentially restricted any American investors, be it institutional individual to invest in tech and AI related sector in the US. That basically just shows this is over. That piece of regulation is not retroactive. So the money that American LPs have committed to Chinese GPs still remains like, okay, to invest.But other than that, once the dry powder runs out, the dry powder that’s raised before earlier this year, when the regulation came into effect, then it’s pretty much over. And a lot of American LPs realize that. And then that’s what makes it difficult for Chinese GPs, for Chinese venture capital firms, becauseIf you are the CIO of a Endowment Fund, you know that, even though you remain committed to about the opportunities in China, but to be honest, DeepSeek only proved that the most potentially lucrative opportunities coming out of China is in the AI and the AI land, which is the various sector that you cannot have any exposure, then what do you do?So then because of this conundrum, that’s what makes it so difficult for Chinese venture capitalists, the GPs, to raise funds from American LPs. just so you know that American LPs traditionally, think before the decoupling that started from around 2021, American LPs actually made up for about half. I mean, there’s no like sort ofconsensus, statistics on this. I’m giving you a number that was given to me by a lot of people in the industry. Before that, before the deep coupling, American LPs make up for like about at least a half in terms of capital raised by Chinese VC firms. And then so that is a big chunk of money that is sort of people know is going to be gone.is gradually dying. So then, you know, to make up for that, you know, we’ve seen Chinese VC firms going to other parts of Asia, Europe, Middle East, to try to, you know, fundraise. And that really hasn’t gone that well for various reasons, right? It takes time to build relationships, to understand the culture.and the thinking behind all these ⁓ LPs and all these big institutions that you are trying to basically ask money for. And then that is sort of on the fundraising part. And on the exit part, for Chinese VCs, the biggest issue right now, the biggest bottleneck, the single, if you ask any Chinese venture capitalists, what is the biggest problem, the difficulties that you are facing?tell you it is because it’s a difficulty to exit. We talked a little bit about the new framework guiding overseas IPOs by Chinese companies. And that has created a significant bottleneck on VC exits. And then so when you are managing a fund, if you can exit and then return capital to investors and then show it in your DPI,then obviously it’s very hard for you to show that this is what I’ve done so that you can raise a new fund. And on top of that, we still have like Ant. Let’s not forget about Ant. I was recently in a conversation with an LP investor in the US who started investing in the China VC space fromlike the 2000s, so pretty early on, right? He was telling me that, you know, almost every big American, you know, endowment or pension fund is still remains like locked in Ant. And without that being unlocked, without some kind of exit in Ant, it also really affects the appetite for this, for them to continue to invest.in the China VC space. I think we just need to see, I know that in terms of the IPO pipeline that we’ve seen, for example, last year, there was a trail of autonomous driving companies being approved to list in the US or Hong Kong, but that’s just not enough, right? We still haven’t seen like say a billion dollar sized IPO.from the tech space. We just haven’t seen that yet. And I think with the absence of clear sign of the IPO pipeline being cleared up without clear sign of a sizable exit, a sizable IPO, sizable meaning at least a billion dollar, it’s just going to be very difficult for Chinese GPs.Grace Shao (51:09)Yeah, I know you actually cover quite extensively when you’re still with the journal. It’s probably like your main story for the last year where you had the journal read before you moved over to information. What are some potential scenarios actually on that point? Like, you know, like you said, this year in Hong Kong, it’s supposedly Hong Kong Exchange has hit like the most filings since like pre-COVID, right? But majority are relatively small.AI or tech companies, they’re not making international headlines. The last one that was really like a major international like candy, eye candy was really Ant and where are we kind of at with that? I mean, this is a bit off topic, but I just find it so fascinating they brought it up.Jing Yang (51:49)Yeah, I mean, just a quick thought on the record amount of filings we see in Hong Kong. It’s basically a lot of smaller companies as well as a lot of companies are already listing in a share market and doing a share listing. With the H-share listings of say ATL and SF, it’s just something that for the moment that Chinese regulators are more amenable to.But in terms of your end question, I think what I have come to believe, and I’m willing to be proven wrong because I’m really only just a journalist, but I think if the NIP were to come back one day, will most unlikely come back in its original form, if that makes sense.It was supposed to be the world’s largest IPO ever. They were on the cusp of raising $35 billion. I just don’t think when Ant finally was going to IPO again, it will come back in that exact shape or form.Grace Shao (52:46)Fair enough. think the company has restructured quite a bit as well, right? Since then. So I wanted to kind of get a sense. I’m not familiar with the space, but are R &B denominated VC funds versus what we were just talking about predominantly being like USC denominated VC funds in China? Are the incentives different for their funds? And do they kind of invest in different kinds of companies? Or are they actually fighting for the same deals?Jing Yang (53:12)Yeah, so first, the incentives are quite different. It’s really like two quite different worlds, right? Different worlds. So in the RMB world, one of the biggest pockets of money comes from government guidance funds and SOE-related funds. And then these types of LP, they are driven by local GDP growth targets kind of incentives. And then so theninvariably when they write a check to you, they would require you to bring back some of the investment back to their local city or province. Say I write you a 10 million yuan check, then that means that usually there’s a very detailed percentage requirement written saying that you need to bring back, say for example, 1 % of your portfolio investments or you need to invest this much in our province or city.And then that is obviously some can argue not very market driven. That is the reality. And I will probably also just play a bit of a devil’s advocate if you are asking for money from a local government related fund, then obviously that is sort of the bottom line. That is the profitability, so to speak, that they are thinking about.Other than that part of the difference, the other difference is that, it’s funny though, like say, know, for a while you don’t see the R &B world and the UST world sort of overlap. For quite a long time, it would just like coexist and then they know that they go into, they all have their own strengths and then they do quite different types of deals or do different types of...going to different types of sectors. However, since this USD funding drought that we have just talked about, what do you see is that some of the Chinese VC firms that were only raising in USD also have started exploring raising our yuan denominated funds. And then there are mostly two reasons behind that phenomenon. First is obviously driven byyou know, the USD funding. Second reason though is also because of the geopolitics made it as we talked about impossible for a lot of American dollar investor to invest in social sectors. So that some of the venture capitalists in China realize there are sectors, for example, semiconductor, right? That is a sector that has just become like not possible.⁓ for ⁓ USD funding to get any exposure in, then if they want to get exposure, if they want to invest in that sector, the only way to do that is through raising our UN-denominated fund and invest. So that’s why you see the two walls that previously just co-exist in parallel now sort of start to overlap.Grace Shao (55:57)Thanks, Zach, lay out.Jing Yang (56:06)love it.Grace Shao (56:07)Thank you for that explanation. Actually, that’s really helpful for even me to understand. think I was also noticing this trend. I was like, why are all these former USD denominated funds, like primarily now actually raising RMB funds, but it’s up to your point to make sense. You know, the Manus AI agent company was probably the most high profile Chinese AI company that received USD VC money. If anything, the only one, right?this past year, what do make of that? And you know, that they moved their operations from Shanghai to Singapore, you know, where do you see this? Is this like the start of a new opening or like a new way for Chinese companies to go find US investment? Or do you think this is like a one-off thing because they did actually receive scrutiny from the US, not so much actually attention from the Chinese that it doesn’t seem like, at least not publicly. So what do you make of that?Jing Yang (56:56)Well, ⁓ Manus is just the latest example, right? But if we look at, if you remember a company, Heijin, that also gained some traction about two years ago, think what we’ve seen that what all these companies, all these startups represent is this new generation of Chinese ⁓ founders and builders, right? They were born and raised in China, some of them educated in China, some of them a bit of China and elsewhere.But they all grew up with that global vision, global view that they want to compete at a global level. However, it’s just so unfortunate that it’s not possible anymore for a company to even try to succeed in both China and global ex-China if you are building a consumer-facing product. That’s just the reality these days. So then realizing that, I think a lot of them now just, you know, if I have to choose China or global ex-China, then...Some choose the former and some choose the latter. And then for those that have chosen the latter, we’ve seen example exemplified by metas, by hygiene and many others.Grace Shao (57:59)I think just one last question, which is if you look ahead three to five years, we can’t predict a future, but what do you think will be the next big story in the China AI tech space?Jing Yang (58:07)Yeah, that is a very difficult question. if I were to try to answer that, would say just follow. I was in the conversation with someone else the other day, know, like, what is a Chinese startup these days? Like, how do you define that? The definition has to change and it has already started to change, right? A company like Manus, right? A company like HeyGen, a company like, you know, GenSpark.Genspar is slightly different than Manus. They actually incorporated as a company in California, but they are funded by two Chinese entrepreneurs who used to work in China. So I think what is really worth watching is these builders and these entrepreneurs. Chinese entrepreneurs, they are definitely going to make waves in the⁓ AI and tech space globally, not just in China, right?Grace Shao (59:02)Yeah, I think we’ve even seen that shift. Remember when Zoom went IPO and everyone was freaking out? Oh my god, it’s a Chinese company, but he’s like an American Chinese. Well, I think he goes naturalized. And just that kind of mentality completely shifted. now, especially in the AI era, more than 50 % of AI researchers are essentially of Chinese descent. So how do you then define it, right? That’s a really interesting take. I really, really appreciate your time. I know you said you had a hard cut off today.Thank you so much, Jing.Jing Yang (59:30)Yeah, thank you so much for having me.AI Proem is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to AI Proem at aiproem.substack.com/subscribe

In this episode, Wency delves into the details of her journey, from working at a leading Chinese-language tech publication to joining an international VC firm, and now writing about China’s AI and tech ecosystem for a global audience. Wency provides a holistic view of the different players from the LLM startups to the leaders in consumer applications and to the rising domestic Nvidia challengers. She is also someone who’s really on the ground and plugged in with the startup community in China, so she sheds light on the real hustle culture and dichotomy of “lying flat” vs. “involution”.Highlighting the differences between tech events in China and the US. She dissects the discourse in China’s youth today, where two seemingly contrasting sentiments coexist. China’s generational gap feels like it happens every five years because of the vast difference of the speed of technology development. Comparing the differences between tech events in China and the US, the invisible hand of the local governments encourages innovation and commerce. She then also addresses misconceptions about Chinese startups and emphasizes the importance of understanding the unique dynamics of the Chinese tech landscape and how Chinese founders still see the US as the benchmark. And ending the conversation with a personal reflection of how her father had never dreamed of the economic growth and personal gains he would experience in a lifetime, a look into the average Chinese man’s sentiment towards the country’s technological development, and showcasing her collection of Labubus.Spotify link here.In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.Chapters00:00 Wency’s Journey from VR to China Journalism to VC and to International Journalism06:48 The Evolution of China Tech from 2020-2025: Trends and Change11:51 Deep Dive into AI: Observations and Major Players23:19 China’s Youth: “Lying Flat,” Involution and the Hustle Culture30:15 Generational Perspectives on Success32:14 The Mission-Driven Entrepreneurial Spirit34:58 Contrasting Global Tech Events38:12 Misconceptions About Chinese Startups41:13 Trade Policies Affecting The Average Business and Average Person43:51 Perceptions of Tech Competition with the West47:58 The Future of Technology and SocietyAI Proem is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to AI Proem at aiproem.substack.com/subscribe

For long-term China tech investors and journalists, she needs no introduction. Rui Ma is a leading voice in the space, a trusted China internet analyst. She is the founder of TechBuzz China. She began her career in traditional finance and then early-stage investing. In recent years, she has advised AI companies and established herself as the founder of an AI school to prepare the next generation for the AI era.In this conversation, we discuss the evolution of the Chinese tech ecosystem and the current trends in AI investment - her bullish view on EVs and robotics. Rui shares insights on the major players in the industry and common misconceptions about Chinese tech. Based in the Bay Area, Rui travels to China regularly, bringing a nuanced understanding of the two worlds and bridging that information gap. The conversation also touches on cultural attitudes towards technology and the future outlook for AI in China. Finally, she discusses her unconventional approach to introducing screens, technology, and now AI to her children.Spotify link here.00:00 Journey from Finance to Tech and AI02:24 Evolution of the Chinese Podcasting Landscape - going on Acquired05:43 Investment Trends in EVs, AI and Robotics08:10 Results > Subsidies10:50 AI bubble?13:34 WAIC: Willingness to pay for LLMs isn't there27:48 Robotics at the front and center31:35 Energy Innovations and Geopolitical Implications35:07 Global Interest in Chinese Technology - Esp. from India40:01 Understanding Cost and Value in Technology Adoption44:42 Unconventional attitude towards technology and educationIn today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.AI Proem is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to AI Proem at aiproem.substack.com/subscribe

Joining me today is Diana Wu David, Director of Futures at ServiceNow. Diana is ranked number #2 futurist in the world by Global Gurus. She ran her own business for nearly a decade called Future Proof Lab where she worked with C-suite executives and boards to help them create future-focused, resilient organizations. She is an expert in guiding professionals and leaders in this ever-evolving business landscape, leveraging strategic foresight to turn uncertainty into a competitive advantage.In this conversation, Grace and Diana explore the transformative impact of AI on the future of work, discussing how organizations can adapt to this change. They delve into the challenges and opportunities presented by AI, the importance of measuring its effectiveness, and the cultural differences in technology adoption between the US and Asia. Diana emphasizes the need for a shift in organizational structures and the importance of preparing future generations for a rapidly evolving workforce. The discussion also touches on the necessity of fostering critical thinking and creativity in education to equip individuals for the future.Spotify link here.In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.Chapters00:00 Navigating the Information Overload04:04 The Future of Work and AI Integration08:54 AI's Impact on Productivity and Organizational Structure14:48 Measuring AI Effectiveness in Enterprises19:41 The Evolution of Work and Education29:40 Cultural Perspectives on Technology Adoption39:24 Unconventional Views on the Future of Work Get full access to AI Proem at aiproem.substack.com/subscribe

Dr. Brian Wong is a HKU-100 Assistant Professor in Philosophy at the University of Hong Kong. His research examines the ethics and dynamics of authoritarian regimes and their foreign policies, historical and colonial injustices, and the intersection of geopolitics, political and moral philosophy, and technology. At HKU, he serves as a Fellow at the Centre on Contemporary China and the World, sits on the Steering Committee for the Hong Kong Ethics Lab, and advises the Interdisciplinary Dynamics: Ethics, AI, and Society at the Institute of Data Science. A Rhodes Scholar and Kwok Scholar, Brian holds a DPhil in Politics from the University of Oxford.In this conversation, Dr. Brian Wong discusses his research at the intersection of moral philosophy, technology, and geopolitics. He emphasizes the importance of understanding AI's impact on employment and the need for academia to adapt to teach AI effectively. Dr. Wong also explores China's tech diplomacy, highlighting its focus on self-sufficiency and global partnerships. He argues that Hong Kong's unique status is crucial for its role in global tech and governance, and he concludes with a reflection on the future of technology and humanity, stressing the importance of human values in the tech race.Spotify link here.In today’s world, there’s no shortage of information. Knowledge is abundant, perspectives are everywhere. But true insight doesn’t come from access alone—it comes from differentiated understanding. It’s the ability to piece together scattered signals, cut through the noise and clutter, and form a clear, original perspective on a situation, a trend, a business, or a person. That’s what makes understanding powerful.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently.For more information on the podcast series, see here.Chapters00:00 Introduction to Dr. Brian Wong's Research02:47 Exploring the Intersection of Philosophy, Technology, and Geopolitics10:38 Chinese Tech Diplomacy and AI Development21:39 China's Global Tech Goals and Misconceptions36:16 Innovation in the Chinese Ecosystem41:20 The Impact of AI on Workforce Dynamics52:16 Navigating AI in Education57:21 Hong Kong's Unique Role in the Global Landscape01:06:43 The Future of Technology and HumanityAI Proem is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Get full access to AI Proem at aiproem.substack.com/subscribe

Today we speak to Cameron Johnson, who delves into the complexities of global supply chains, particularly focusing on the dynamics between the US and China. We explore the implications of trade policies, the evolving role of Asian countries in manufacturing, and the intersection of AI and technology with supply chain management. Cameron shares insights from his extensive experience in the field, highlighting the multifaceted nature of supply chains and the importance of understanding the broader ecosystem that supports them. Furthermore, we discuss the intricate relationship between government regulations, robotics, and supply chains in China. And finally, Cameron shares his conviction that well-managed supply chains can foster peace among nations.Every episode, I bring in a guest with a unique point of view on a critical matter, phenomenon, or business trend—someone who can help us see things differently. Get full access to AI Proem at aiproem.substack.com/subscribe