A (3:16)

And I guess for China it's not just about, I suppose in the west we think of it as a trade off between sort of beneficial exchange of information and freedom and popular discussion versus misinformation, disinformation and hate speech. But then in China, an element where as an authoritarian regime they, they aren't entirely happy with, with just open discussion in general, regardless of whether it's true or not. Right. So that's the kind of additional dimension that they face.

C (3:43)

I don't. So Ting, I think you should come in here as well. But my understanding is also that it's very much up to the, these societies and of course also, you know, various actors which are political actors. So China obviously the Chinese Communist Party, because it's a one party state is extremely important. But then also societal actors come in, also private actors like the tech companies, they also come in with their understanding of where are the boundaries for freedom of speech. And so I think for China the answer has always been very state driven, very party driven. Whereby kind of the bottom line or the party line would be There are certain topics that Xinjiang, but also of course many others that are considered to be destabilizing while then in other societies this would be considered to not necessarily be stabilizing. So for example, I lived for 10 years in the Netherlands and in the Netherlands, culturally there's an understanding that if you were going to suppress those voices, that would be destabilizing because then people continue to have grievances but they're not discussed openly. So in order to stabilize society, you have to open up more freedom of speech to discuss issues. So I think it's also a value driven answer. But I think different actors across the world, globally right now are trying to rethink where the boundaries and their societies are.

A (5:24)

So you sort of start by, you know, as you mentioned, like in China, certainly especially compared to, you know, Europe or the US it's much more, much heavier government influence and the bottom line is kind of more driven by the government. But you sort of set, you set up a framework you characterize as popular corporatism, which you distinguish from a kind of a pure command and control. So you know, I think you're probably your average, let's say American senator would say if you ask them about China, they would say that China's just, you know, they control everything and you know, there's surveillance on everything and the party is pervasive and they decide exactly what speech happens and there's all sorts of censorship and that's not entirely untrue. But, but tell me more about like what, what you found in your research and what this popular corporatism idea entails.

B (6:17)

I think I can, I can jump in here. I would say I really like when you mentioned people would outside of China they would like see China as, you know, some something like Black Mirror. So you can have for example social credit system. You will have them, everyone have some sort of chips in their brain or whatever and then they can rate it by their social status and the state will be able to know your score and then based on your score to allocate your resources and to decide whether you have access to jobs or loans, for example. But I think the reality is not true in China. It's much more complex. I think that's what we want to argue here. And also our evidence suggests because the state will have to rely on the company to build those technology, put it that way. They will rely on the company to able to basically only the company have the expertise, have the which to a large audience have the database and also have the technology to build everything. But then state will know that they won't be able to do it themselves, the state, the government. So they will have to work with the company and then for the company in order to asset those data, for example, if you're able to get some scores of information about a citizen, you will need to have citizen participation on those platform to provide those data. And that's also part of their business model. So in a way that we are seeing the company as some sort of in between the states and the citizen. And citizens will be able to voluntarily contribute some data in order to participate in the platform. And here the company will be able to sort of balance what they want from them, but they also can balance the user's requests from the platform. So basically you can see that as a balancing act of the company. Danny? Maybe. Yeah.

C (8:03)

If I can come in here. I think what Ting is emphasizing are bottom up mechanisms that really differ in our framework, which we call popular corporatism in the book and it's called popular corporatism because we want to emphasize that of course the state ultimately in this partnership between the state and these tech companies is in the driver's seat. So yes, it's true. China obviously it's an authoritarian state. Obviously the state has very important, both coercive but also financial mechanisms through which the state can set boundaries but then at the same time in the past in what we call command and control, which would be sort of the stereotypical or dominant understanding of how digital governance works in China. In this command and control understanding, companies are just intermediaries where they're essentially just enacting what the states goals or motivations are. But what we are finding based on empirical evidence, such as the social credit system. I think Ting just started out with this example of the social credit system is that these bottom up and mechanisms and the dependency of the state on these platforms and on the companies, but also the dependency of the companies on user participation in order to increase their reach into society, but also in order to update, constantly update their knowledge through data, this dependency really is very critical in order to understand also the constraints that the government has and also acts of resistance that we also lay out in the book. So where both citizens, but also companies are showing resistance and are not just simply sort of complying to state demands. And so that's where corporatism comes in as an understanding that kind of the party is of course kind of like more like a CEO in their relationship with the companies. But then it's also popular because you need citizen participation in order for making this entire system work.

A (10:35)

Right. So just to sketch things out. So in your book you talk about two different arenas basically. So one first section of the book focuses on social media. So WeChat run by the company Tencent and Weibo run by Sina, it's kind of the main focus. And then you shift over to the social credit system with Alibaba as kind of the main player there as well as, as well as Tencent. Why don't we, since, since Ting introduced the social credit system, why don't we start there to sort of make it concrete, you know, talk about like the policies sort of that you might, you know, that again say, you know, an American who's unfamiliar with the system or you know, Saw, Saw, Black Mirror might think what's going on? And then what's kind of actually going on and how the, how the bottom up pressures are actually companies to shape what the companies can do and then also affect how the companies, you know, when they choose to push back against the, against government policies.

B (11:37)

Yeah, that's. Well, I'm thinking about how to structure my answer because that could be quick because I think I first need to introduce the whole system. So I think at the moment, as far as we know, I mean at least in scholarly discussion, social credit system is quite often considered some sort of political control or Political surveillance system. So mostly associated with for example local government credit. I think most people basically that's a narrow sense of social credit system. And we also, we call that basically in our definition that is the political part of the social credit system or political aspect, but we also have the commercial aspect. And commercial aspect basically are those that, for example Sesame Credit or Tencent Credit. I think Tencent Credit is now called Richard Point. So it used to be Tencent Credit, but later on change it to WeChatPoint. But nonetheless, regardless of whatever example we provide here, generally they are those commercial credit built by commercial company, but they are helping the government, especially the Central bank of China to build these commercial credit system that's commonly known outside of China. For example, in the U.S. i think you have something similar, the sort of credit system in Germany, I can't remember.

C (12:48)

It'S called Fico School, it's called Shufa in Germany. Yes, it's a financial credit system.

B (12:54)

Yes, exactly. So I think in China we also that we in our definition or when we see there, if you look at the evidence, we actually also have the commercial credit system that's similar to, you know, whatever that you have outside of China. And so basically social credit system broadly in China we have two aspects. The political aspect that's closely tied to for example local government credit, but we also have the commercial one that is built by commercial company and that was aimed to be able to provide some evaluation of people's financial credit worthiness, for example. So that's more commercial. But then the key difference from outside of China is these two systems somehow is at least our evidence suggests they have potential to be linked. But then it's not really like what we have portrayed in Black Mirror. But you use the commercial system to somehow link to the political one and to enable political control. And what we see more is how the company, because of the expertise in building the commercial credit system, they are also helping the local government to able to build their own credit system or somehow merge some of those data together. At the moment it's more about behavioral compliance rather than any political control. We haven't seen any political control element yet. I think Danny. Right. Our evidence haven't suggest any of those for now.

C (14:09)

Yeah, maybe. I think one of the greatest fears has always been that for example, Tencent has a lot of user data from WeChat, which is used by, according to our data, which is nationally representative, 99% of all Chinese Internet users, that's a huge Number are on WeChat. And so that's a lot. I mean, that's a huge monopoly. And the fear has always been from the outside that those data would be linked. And then of course, political attitudes and, and sort of political behavior that is organized on WeChat would be merged with, for example, Tencent Pay. But according to the empirical evidence that we compiled in the book, which is largely based on interviews, qualitative interviews, but also nationally representative survey, we couldn't find any evidence for such links between those data. But there's also, the second fear has always been that there's one national credit, social credit system where the central party state then controls it all. And because of the way this is set up, and I think Ting Ting can explain that a little bit better, there's actually competition between the companies so that they don't have an interest in emerging data. So that kind of leads to a fragmentation, an ongoing fragmentation of the way it's structured.

A (15:48)

Yeah. So the US News Media summary. Yeah. Tends to sort of conflate several things. It sort of alludes to things like Sesame Credit, which are more based on mostly, as I understand it, buying and selling information. Like, did you actually, you know, ship the order that you said you'd ship? And do you pay for things on time? You know, the same kind of stuff that we would consider part of a credit score or maybe your reputation on like ebay, these kind of reputational scores on a platform that, that's kind of one thing. And then there's this idea which, you know, had some, I guess, trials started but never hasn't really, I guess still hasn't really gotten that far of integrating this with more, I guess, like you said, compliance. Like, you know, do you, I mean, I think other instances people have said, like, do you take care of your grandmother? Do you like, say nice things or mean things about Xi Jinping online? Or do you, you know, do you repost memes that could be viewed as SC skeptical? And then maybe somehow your, your score will get lower and lower and then it gets merged with this, you know, idea of a blacklist. And then if your score gets low enough because of your jokes about, you know, your online memes, then suddenly you'll be barred from, you know, existing in the Chinese world entirely. You can't get on an airplane. And there's, so there's all these things kind of get muddled together. So, so what, what actually is, is happening. So there's this that you do mention, there is a local government element of it. So there's the social credit system, which is really, like you said, much More of a credit scoring. I think it has much more pervasive access to information than the credit scoring systems in the US I think they're much more restricted. Whereas if you're on WeChat, Tencent, anything it wants to look at to determine your score, it can do. So that's one thing. But yeah, but tell me more about what is the government involvement or how would the government want to. How did it want to get involved in the social credit system and how did that pushback happen?

B (17:45)

I think at the moment when it comes to sort of integration or you would say the application of this in compliance mostly in legal systems. So basically in carrying out legal. I think it's legal ruling for example and mostly financial relay. So for example it's for the financial disputes and there is a ruling about someone needs someone on the other person's money then they would actually use enforce those with the assistance of e.g. vCHPay or Pensim or Sesame credits. Because you won't be able to buy tickets, you won't be able to especially expensive tickets like first class or business, et cetera. So I think this is at the moment the only place that we see those connection.

A (18:32)

And it's only for actual documented legal. Legal disputes.

B (18:35)

Yes. You have to like actual financial. Normal is financial disputes. So you have to be legally, say legally have those put in a legal document. I think that's the only place we've seen this. And so I think there was something other question, I can't remember so that.

A (18:50)

So that's so far that's the only kind of integration with the political system. And really that's more like. Like you said, it's more integration with the legal system and you know, improving on is like contract enforcement which we generally be view as a normal function of a functioning market society. You need to have that anyway.

B (19:08)

All right. And also I think it's more about why local government will need those commercial companies to help them has to do with the local government credit. Those credit system has a problem of low participation rate and people are not using them and they cannot attract to a wider sort of user base and they will need a company to help them to set them up so they will be able to attract local especially citizen to use them. Because if you have no one to use those app then those credit score has no meaning because you don't have any data. So I think that's also the power we feel of the collaboration at least in some local government.

A (19:44)

What would a local government credit app look like? Is that something they would, they'd be hoping that every business logged into and registered their business disputes on or what would.

B (19:56)

I think it's more like a daily use. You know, at the moment it's very simple. At least based on our book when we were doing our research. It's very, very sort of basic. You know, whether you violate any traffic rules, for example, or whether you borrow the book from public library where they return them. That could suck. That kind of public service, you know, you're complying to public sort of rules or regulation. But very general at the moment.

A (20:21)

And is this the, the, is the issue that there's not users who are, I mean you. I think if I were a user and I was, you know, going to not return my library book, then I'd rather not be in that system. So I wouldn't join anything but it. Okay, so. Okay, so then, but since everyone has to be more or less to function in China now you have to be on, on WeChat. So therefore everyone's on WeChat. And then if they want to. So if they link that with your, let's say your public library record or your speeding tickets or something else, then they can. They have more leverage over. Over people.

B (20:55)

Exactly. Yeah. Yeah, I think that's part of it. Yes, it's part of it. Why they need a company to help. I think what, what we see here is also we see some companies push back because they will worry about their, their sort of user base because if they release too much information to the government, they worry that citizen might move to other platform. So that's also harm their business interests. I think that's also what we mentioned in our book that there is these, you know, this bottom line of our users and company do care about those users.

A (21:22)

Right. And I guess so in this case with this credit system that's really between, between Tencent and Alibaba are kind of the two major competitors for payment systems. So if, if either of them becomes too strict about things that people don't view as legitimate or then the other side. So it's kind of a. Not a race to the bottom, I don't know. But yeah, anyway, there's competitive element that keeps them from wanting to be too strict.

C (21:46)

Yeah, exactly. And so what we also find in the book is that on the one hand it's a strategy also by the government, a more broader strategy. This is not specific to the digital economy, but just in general Chinese economic development, it's a very state driven like the other Asian tigers, China has a very state driven economic development approach. And so it's in the interest, of course, of the government to make sure that there are not too many players around. And they develop policies to also create incentives for certain kinds of companies to develop more. But then at the same time, you also want to have a certain element of competition still amongst these players. And so because it is essentially still a competitive, sort of a limited competitive system that also creates incentives for the companies to respond to what we call the bottom line. And so I think what's important about the bottom line to understand is that it's really not the individual user that has any impact here. So if you as a user in China decide that you find the social credit system to be too intrusive, for example, during COVID in certain parts of China, the social credit system, there were some reports, maybe you read about some of the reports, where it's being used to also enforce the zero COVID policy, right? And if you as an individual user then decide to leave somehow, for example, Sesame Credit, it's not really affecting, of course, the company's behavior, because the individual user's choice in itself doesn't matter so much to the company. But what really matters are these kind of almost like mass movements, so collective action where large numbers of users are suddenly migrating. And so we've seen that, for example, also outside of China, this can happen. When Twitter became X, lots of people moved out of Twitter, moved into blue sky, LinkedIn, you name it. And then of course, it's still to the company to react to those instances. So not obviously there's still a management decision. So as we have seen, in case of X, there was limited impact probably of these mass movements away from X, but it's still, it still figures into these business decisions. And so in China, I think, like in comparison, Chinese users are already adopters and they also tend to be very sensitive towards other users trying.

B (24:38)

They.

C (24:38)

It's kind of like a very curious environment. You know, there's a new product somewhere and then you hear through your social network, hey, you know, there's this new app, why don't you try it out? And so then it becomes kind of, it kind of gains traction very easily. And so in China, even if you have a very established, for example, I can give an example, like Cinablog, for example, was very, very trendy in the early 2000s. Lots of people were on Cina blog. And then at some point, micro blogging became much more trendy. And so now CineBlog still exists, but it's very, very limited in terms of use. And most people have moved on first into microblogs and then later into WeChat. And so those moments are the ones that then really also play a role in the business decisions of the companies.

A (25:39)

Right. So, yeah, so I guess it just in general, in thinking about platform economics, there's always this, you know, there's an important aspect of network externalities where you want to be where everyone else is. But certainly my experience of, I think theory being a little bit overturned by a reality both in the, in the, in China and in the west is seeing actually a lot of these. It's surprising how something like a dominant, what seems like a dominant platform, like you know, Facebook back in the day can really just kind of everyone just loses steam on it and then, yeah, and people are switched to other, other platforms and then their, their business really suffers because people are surprisingly more. More willing to jump to something new than I think the. Our basic theory would. Would suggest. And you're saying in China that's particularly true. People are really excited to, to try something new. And I guess, you know, we see that, you know, it like in the US like switching from, you know, Facebook to, you know, getting bored with Facebook and then everyone's on TikTok or Instagram or whatever else and then, yeah, so that, that creates, creates this competitive pressure.

B (26:42)

Right.

C (26:43)

And so what we see in case of the social credit system is that there it leads to. We cannot obviously, like in any other country in the world, it's very difficult to speak to, you know, CEOs of big tech companies and have an honest conversation with them. So, you know, China is not, is no different than American or, you know, any European other tech companies. So, so we don't know. But in terms of the evidence that we could collect, the evidence really suggests that there has been, of course from the central government, a conversation with Tencent and also Alibaba and also there are several other companies. Ting can explain that even more, but totally, I think eight tech companies, if I remember correctly, Ting is nodding here, that are in that have. Are not exactly competing with each other, but they have sort of social credit scores that are relevant. And so the central government has of course demanded for greater integration of these existing data because all of these players, they have data on certain regional, sort of, it's regional and on certain types of data. And the companies have really resisted and delayed this process to the extent that they actually at some point started to propose their own solution, which still needs to be seen whether it is actually a solution, but it's ting. Do you want to explain the solution since this is very much your expertise.

B (28:35)

I see, yeah. So basically the solution they suggest this is also based on our reading of another scholars legal scholars article. So they basically suggest something called United Agency. So United Credit Score agency. So let's have a United Agency that is led by the central government or you know, and sort of an arm or investment arm of the central government at least is in control by the central government. And then, and then our commercial company can be sort of holding certain share of tech company and in this way they can, in a way that the company is still owned by the basically these sort of United Credit agents is still owned by the government, government still in control. But then company, because they all have a share, then somehow they can negotiate how much data they can share with these credit union agency. And the credit union agency as a single sort of company can own these credit licensing and all these separate. So being part of the board, they will be able to share that license. They will continue to be able to run their session credit, for example, and then tencent credit. So I think this is a very. We see that as a very creative solution to this because in a way you can see these as, you know, at least the government can still feel like they control everything, but then company retains some agency or some sort of control over how much to share what I want to share.

C (29:55)

And we do not know because we. So we could not collect any evidence on how to what extent data is actually being matched through the solution. So various observers and also people we interviewed in China, they said technologically it's impossible because all these different social credit scores come in a different data format. And so technology experts say it's actually impossible to completely merge the data and make it into one huge data set. Merge it into one data set so that you can analyze these data together. So yeah, this is an open question to what extent this is actually really resolving this problem about the fragmentation of the system. I remain skeptical.

A (30:50)

Does this exist? Is it. So has it been created? And it's functioning, but maybe not that well. Is that where it's at?

B (30:56)

Yeah, at the moment I think we got two right. One, I can't remember there was, I mean when we published our book, there was two. I think it's. I can't remember the name. I think the second one is by. The first one is by Han. The second one, there was different company sharing holder of these United Credit Agency. And then the third one, they were news about having a third one as well. So at the moment, we do know there is this Credit United agency, but I don't know whether they're functioning well because there were rumors about all these commercial companies decline to share data, so they do not really have the kind of data that allowed them to fulfill their vision when they were first set up.

C (31:35)

Yeah. So it remains to be seen whether it actually proves to be a viable solution in the end.

E (31:43)

This episode is brought to you by Indeed. Stop waiting around for the perfect candidate. Instead, use Indeed Sponsored Jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate. C According to Indeed data, Sponsored Jobs have four times more applicants than non sponsored jobs. So go build your dream team today with Indeed. Get a $75 sponsored job credit@ Indeed.com podcast. Terms and conditions apply.

F (32:12)

The world moves fast. Your workday even faster. Pitching products, drafting reports, analyzing data. Microsoft 365 Copilot is your AI assistant for work built into Word, Excel, PowerPoint and other Microsoft 365 apps you use, helping you quickly write, analyze, create and summarize so you can cut through clutter and clear a path to your best work. Learn more@Microsoft.com M365 copilot.

D (32:42)

Time is valuable. That's why Lowe's Blueprint takeoffs turn blueprints into quotes faster. Bring us your plans and we'll generate itemized material lists to make quoting easier so you can get back to building. Plus, at the Lowe's Pro desk, you get access to thousands of building materials not sold in store. And when your order is ready, we'll deliver everything to the job site. Improving is easy at Lowe's, Right.

A (33:11)

So both political and technical challenges too. You know, first of all, whether the companies are willing to share information upward to this group. And then also. Yeah. Trying to reconcile that. I mean it seems like it'd be easier in China because at least there's, you know, people have a unique ID number. But I guess if you have. Yeah. How do you compare a certain kind of behavior that. Yeah. That WeChat's keeping track of versus some other set of behavior that Alibaba has and integrate it into a single score. That does seem like that would be. That would pose a lot of challenges for to. To find a sensible way to do that and yeah. And then I imagine the actual data sets are probably organized very differently and so. Huh. So how has this changed over time? I mean, so your big survey was a few years back that a lot of this is based on and, you know, since then, there was kind of this big Internet crackdown post Covid. So do you think, does that change your sense of what's going on or, you know, how. How is the shifting power between, between the firms and the state, especially, you know, with Xi Jinping generally taking a much more controlling approach to everything.

C (34:25)

So what we discovered in the book through our China Internet survey, which was it's, it's sort of the last of a kind. So it used to be the case that in the China field, we were able to use what's called GPS random sampling in order to make sure that we get representative public opinion data, also on migrants who are moving from the countryside to the cities and who are not registered. And this was conducted in 2018, and it's the last survey, at least to our knowledge, that included political questions such as on political trust or how do you see the social credit system system? And so this survey really allowed us to also engage with the implications of the system and how users, first of all, why are users participating? It's obvious that the state is very involved. It's also obvious to a certain extent that the companies have their own interests, their various scandals on privacy leaks and so forth. So why is that? And so through this survey, we discovered that there is kind of a, in China, we wouldn't use that term liberal, but there is kind of a liberal bias in the sort of majority views of Chinese Internet users to the extent that there is to the contrary of what many people outside of China argue that. Or also oftentimes even in China, you also hear these statements, Chinese citizens or Chinese people just don't care so much about privacy. And that's why in China this works differently and it's okay for citizens that they don't have any privacy and so forth. So we can't confirm these claims based on our data. We find that, no, to the contrary, if you ask people about what they are doing, if you ask about, you know, what they're doing, what kind of privacy settings they have on their phones and so forth, you discover that the vast majority of Chinese Internet users actually does take action, or they do take action in order to protect their privacy online, and that they also, there's also a preference or priority for less censorship, sort of more freedom. For, of course, the term freedom, I'm using an English term here, so I don't want to interpret too much into it, but definitely, if you stick closer to the way you would express this in China would be having more space to voice criticism. And so These preferences, in my view, have not changed because this is not a change. This has to do with values. And these kinds of values usually change very slowly. But of course, since we could collect the data, which was most of the data, the whole book is about comparing the changes between Hu Jintao and Xi Jinping. And then we conducted most of our data before COVID There is of course the question to what extent have, for example, has the antitrust campaign that was launched in 2020, to what extent has that really changed the dynamics? And so in our view, in my view, but think you should of course also come in here. But I wouldn't say that the preference of the users has really changed. But of course the started to realize through some statements. The antitrust campaign started with a speech made in Shanghai at the Bun by Jack Ma, where it became very obvious he made public statements that were suggesting the power openly. He was speaking about the power of Alibaba and was also speaking negatively about sort of Chinese central bank and sort of Chinese bureaucracy. And so we see through the antitrust campaign, kind of, in a way it confirms what we find in the book that the power of Alibaba and other tech firms and that dependence of the Chinese state on these firms and in the antitrust campaign, an attempt to clamp down on them and to show them their place. But then also we're looking into procurement data in order to expose this dependency. And so these procurement data, for example.

B (39:47)

They show.

C (39:50)

This is publicly available data that the Chinese state publicizes. In order to be more transparent and to cope with potential corruption. And so in those agreements, the government units that give contracts towards various tech companies, they need to justify why they choose Alibaba or why they choose to work with Tencent or Sina and so forth. And so through these procurement data, we can also, in the book we also look into the sort of the post Covid period and what happened after 2020. What you see very clearly is that amongst the big tech companies in China, Alibaba suffered the most. So we think that in part is related to this being sparked by this personal speech. So it's kind of personal in a way, that's my interpretation of it. But also you can see, for example, Baidu, in comparison to other companies, they continue to have these procurement agreements to a larger extent compared to other companies. But you can also see that it kind of, after a few years, it kind of goes back to what it was before. And so this is of course, this is not new to anybody who studies campaigning in China, because campaigns they tend to not last forever. They only tend to last for the time that the Chinese state or various, if it's regional, it can be a regional state but it can also be the central government to the extent that there is a lot of attention being devoted towards those issues. So we don't think that the popular corporatists partnership and relationship that we discovered was fundamentally changed even during the antitrust campaign. Ting. Do you want to also come in here on sort of changes under Xi Jinping?

B (41:58)

Yeah, I think I just want to add on to what Danny mentioned. I think something I also find very interesting is we do find a parallel to, for example AI government in China. So we can see that popular copulism can actually explain what's going on to AI. You can see that artificial intelligence especially for example, let's say Gen AI something like deep sea for example. I mean this also is participatory platform. You can see them as similar to what we have in our book participatory platform. You need data, it's not just amount of data, it's a good quality data. And if you need those data you need citizens input. And we see exactly the same as we find in our book. You see State want to control those company and you see state published, I think Cyber CAC published this new rule asking all gen AI have to follow socialist value. So this is typical control. But on the other hand you also see Chinese government have set up these in the past when we was writing our book they have Internet fund, now we have AI fund. All this fund is meant to invest in all those key players and to promote AI development, give them more money to develop AI technology, et cetera. And then on the other you see these sort of, you can see that as some sort of top down between top down control and bottom influence, right between the partnership between state and company. But on the other side you can see company because they require citizens input, users input into the platform to help them improve the output of for example Genai. They still have to balance, try kind of balance the bottom influence of citizen. So I think pretty much we are very pleased to find that actually our model can still explain what's going on not just to digital government but also to the cutting edge technology like AI.

A (43:45)

Right? So the commonality is. So I guess the commonality is that in both situations you need participation from citizens. So if you make things too closed or repressive or boring or not useful, then they just opt out or. Well I guess when there's multiple competing systems and there's A competition between them, so they could opt from one to the other, but they could also just opt out and like, not, not get involved at all. And so even, you know, maybe the government recognizes that even if they were directly running these things, they, they wouldn't be able to, you know, completely impose their will. Because, you know, if you, yeah, if you put, if you press too hard, then, then no one, no one will engage with it at all. And you mentioned we haven't talked much about the social media, but certainly very much the same thing there. Right. You, I think you mentioned, you know, with Weibo versus WeChat, that Weibo kind of declined as it became, as the censorship tightened up, so it was no longer as much of an interesting forum for people, and they just left. And then that the government loses the value they might have gotten from those conversations.

C (44:54)

Exactly. And what you also find is that interestingly, these big. So the players that do have this partnership with Chinese central government, but also local governments through a business relationship, I think we forgot to mention that earlier, this is very important, that this dependency also comes about by these government units to become clients for the platforms. And so it is through these procurement notices that we track sort of the financial relationship and the financial partnership and dependency between state and corporations. And what you find is that those companies, like Sina, for example, even though Cina Weibo only has about 20% of Internet users, so compared to WeChat, I mentioned this before, which has a complete monopoly, so everybody's on WeChat. Right. And so Cina Wayboard, you know, 20% of Chinese Internet users are in Cina Wayboard. But nevertheless, those user data are still extremely important to local governments because the people who are on Sina Weibo, they tend to also talk about issues that are considered to be destabilizing. So there may be oftentimes there's a correlation between certain trends and certain criticism and certain topics on Sina Weibo and actual protests happening. And so in order to avoid these protests, there are, There are.

B (46:41)

There's a.

C (46:43)

Many local governments are actually, for example, in Fujian, you know, there's. There's an example where the city government of Fujian asked Sina Weibo to help to promote the accomplishments of the Chinese Communist Party of, of the past in order to sort of promote a positive vision via the platform. And then through this relationship, of course, then the marketing department of Sina Weibo can introduce certain ideas about how this could be done and so forth. And so if you have that kind of. We call this an insider relationship. So if you have that kind of insider relationship with the state, you actually have more leeway as a platform to not censor as much. So that's kind of surprising, Right. So even though there's more attention towards your user base and your platform, nevertheless you gain a little bit of leverage in order to just not enforce so much. And so in comparison to then other players who don't have that sort of insider relationship, you actually can, can, you can, you can make sure that your users are a bit happier because they can talk a little bit more freely.

B (48:07)

If that makes sense.

A (48:08)

So it's kind of a trusted partner role. So rather than being, if you're more arm's length, then you have to be more cautious to avoid upsetting people in the government. But if the government knows kind of is more closely interlinked with you, then then they'll have a better sense of what you're doing and that be more, let you have a little bit more. More flexibility.

C (48:27)

Exactly.

A (48:29)

Great. Well, we're just about out of time. There's obviously a lot, a lot more in the book we could talk about. I'm sure we could. Hopefully we'll be able to meet up in person someday soon and have a longer conversation. But why don't we close. Why don't you guys each let me know what else are you working on now? What's your next project? Together or separately? Ting, you go ahead.

B (48:52)

Right. So actually at the moment I'm working on. So as I mentioned earlier, I want to try to say how our model can be applied to AI government. So currently I'm working on AI government. So try to see how the Chinese government manages to develop AI also to control or control the harm that can potentially create by AI. But I think this is more like a bigger project. So at the moment I'm more specifically looking into, I think earlier we mentioned about these different values in the west and in the East. So at the moment a small project I'm working on this part of the bigger project on AI governance in China is to look into how this AI system generally a gen AI chatbots that create, using different data and also with different rules, regulation and context, how they might interpret different values, how they might interpret value differently. So I think this is at the moment. So I have a fancy title for this. I think I would say the project's title is something about the political ideology of AI systems.

A (49:53)

So not the ideology of the people creating the systems, but the ideology that sort of becomes within the system, manifests in the system.

B (50:00)

We're thinking about is a discurse pattern of the. I'm not saying that AI has, you know, has ideology, but more like how it's been manifest throughout the process as an output. Yeah.

C (50:11)

And I'm curious is. Is your answer that it's just data? So it's because. Because there's a bias in the data, then the system learns through politically ideological data or are you also looking into other hypotheses?

B (50:32)

At the moment I think I'm looking into. I mean data is part of it. I think people working on LLMs, I think we all have this puzzle whether they learn this from the data or whether they pick this up later through the reinforced learning with human feedback features. So all these platforms, they will hire coder to actually reward the behavior of the system. So do they pick it up at that stage or earlier through the data or if it's through data, which stage of the data? The very early as a foundational model or later when you have better data, where do you pick up those political bias values or different values, put it that way. So I think this is something. That's what I mentioned. This is a much bigger project because at the end I want to be able to text, you know, at which stage is it, you know, which stage contribute to this? Maybe I will never able to find an answer because you can't really separate them.

A (51:24)

Wow, that's a really cool project. Yeah. Looking forward to hearing more about what you find out as it progresses. Danny, how about you?

C (51:32)

Yeah, I'm also fascinated by this. So it's always. It's wonderful to work with Ting. So I'm really looking forward to continuing this conversation as so I currently, I'm really fascinated by what's happening in Europe in terms of this new approach of the European Union towards governing or resolving the digital dilemma in Europe. Because there really has been a shift.

A (52:03)

From.

C (52:07)

Kind of a consensus has been built inside the European Union that the previous model, which it used to be just a global understanding of a governance of tech companies, where tech companies tend to have a lot of autonomy and that used to be a global one, existed in the us, in Europe, in most places in the world. But now in Europe there has been a real shift where consensus has been built that this is not helping to address the problems and it's really destabilizing liberal democracy. And so now we have a series of new digital regulations in Europe, including most important ones are digital. Of course the AI act has received a lot of attention, but more important than the AI act is the Digital Markets act. And The Digital Services Act. And so what's so fascinating to me is that of course in Europe, because Europe doesn't have any strong own homegrown tech companies. Sorry, no, no offense to any, any, any, any, any audience, any members in the audience. But I would say it's like in comparison to the US And China, Europe, in terms of its own digital innovation, it's not its strength. Right, so then, so then kind of Europe is, is, is of course making it this weakness into a strength. I also call this digital place digital aikido by, you know, regulating or then having also the freedom to regulate both China or Chinese tech companies and American tech companies. And so what I find. So I'm, I'm, I'm now really fascinated by how both American and Chinese tech companies are, are reacting to these new regulations and sort of, at least some of these tech companies, like TikTok for example, and TikTok of course always claims that it's not really a Chinese tech company. And you know, so, so there is also the question, what is a Chinese tech company? But TikTok, TikTok really tries to be the kind of the good guy in comparison to the, you know, other American counterparts are resisting these new regulations and playing along and really trying to comply. And it's really fascinating to me and there's also another underlying question here, which is to what extent can regulations actually then lead to compliance of tech companies that are very closely tied to other states? And so I'm currently really fascinated by these recent developments and, and we'll see what the answer, the empirical answer will be to this question.

A (55:10)

Okay, well, it's been great talking to you and this really exciting work you guys have coming up. Again, the name of the book is Governing Digital China. Definitely encourage everyone to get a copy and get a copy in their library and read it. Really important for understanding, well, China's digital economy, which of course these days is the economy. So if you don't understand that, you don't understand anything. So really great book and thanks again to you both for talking with us today.

C (55:39)

Thank you so much for having us, Peter. It's been great to have this conversation and really fun and hope you also enjoyed reading the book.

B (55:48)

Thank you, Peter.

A (55:56)

Foreign.

E (56:03)

Starting a business can seem like a daunting task unless you have a partner like Shopify. They have the tools you need to start and grow your business. From designing a website to marketing, to selling and beyond, Shopify can help with everything you need. There's a reason millions of companies like Mattel, Heinz and allbirds Continue to trust and use them. With Shopify on your side, turn your big business idea into Sign up for your $1 per month trial@shopify.com specialoffer.