Does Banning TikTok Make Sense?

Shoshana Wodinsky, Daniel Greene


Paris Marx is joined by Shoshana Wodinsky to discuss the unconvincing arguments being made for a TikTok ban in the United States, then by Daniel Greene to explore how the turn against Chinese technology signals a shift in US policy on the internet and technology.


Shoshana Wodinsky is a freelance reporter, previously at Marketwatch and Gizmodo. She writes the Tubes newsletter. Daniel Greene is an assistant professor at the University of Maryland’s College of Information Studies and the author of The Promise of Access: Technology, Inequality, and the Political Economy of Hope. Follow Shoshana on Twitter at @swodinsky and Daniel at @Greene_DM.

Support the show

Venture capitalists aren’t funding critical analysis of the tech industry — that’s why the show relies on listener support.

Become a supporter on Patreon to ensure the show can keep promoting critical tech perspectives. That will also get you access to the Discord chat, a shoutout on the show, some stickers, and more!



Paris Marx: Shoshana, welcome back to Tech Won’t Save Us.

Shoshana Wodinsky: Happy to be here. Thanks for having me on.

PM: I’m very happy to have you on the show because obviously, all of this stuff has been going on recently with TikTok, this is something that is not entirely new. We saw it during the Trump campaign and you were writing about it during that time when you were working at Gizmodo. But now, the Biden administration and the US Congress have brought this back. TikTok, is back in the news. The government is talking about potentially banning it. It’s already been banned from government devices in many states across the United States, I believe federally as well, and oof course, many other Western countries have done the same and followed suit. Now on March 23rd, TikTok CEO was in front of the US Congress to answer questions. I just want to start us off with the big picture and then we can zoom in on specific aspects of this, but what were your thoughts on the hearing and the larger US campaign that is being waged against TikTok?

SW: How long do you have? I wrote several thousand words about the first hour of the hearing alone. As it was going on, I was talking to a few other reporters in the space about this. I said that it felt kind of like five different tech hearings, all cobbled together in some kind of like nightmarish golem. Because the questions being asked, they kind of related to each other — they were all about apps, they were all about phones, all about tech. But it sounded sometimes like they were talking about completely different companies. When they were talking about TikTok, they were getting a lot of basic stuff wrong half the time.

Look, I say that I’m an optimist about this stuff, when it comes to tech regulation. We’re seeing great moves being made against Google, for example, which is facing several anti-trust suits by dozens of states. But then you watch something like this, the same Congress people that I’ve been cheering on with the Google stuff these past few months — one of the questions was literally about whether the app tracks people’s eye movements for either targeted ad purposes or something like that. I was listening to that, and I was like: Where did you get this idea? I wasn’t even mad. I was just confused by a lot of the hearing.

PM: I would say when I was reading your newsletter, I was also frequently very confused. Because you not only had your thoughts on these different things that the Congress people were saying, but you also took out of the transcript, some of these different things, and really wild assertions that in some cases had some kind of grounding, but in other cases just seem completely out there.

SW: Wild assertions that didn’t let him get a word in, edgewise. Like they would say something more than once. A Congressperson would say something completely out of left field and very much not true, and then Shou would just be like: Am I allowed to respond to this? And they would say: No, you have to talk to them. I came away from that hearing, feeling bad for the guy. This is his first ever tech hearing, like against Zuckerberg or Pichai or Dorsey, and I don’t think Congress people have been this mad in a long time at one of these hearings. I think that’s owing to them being mad, like they picked the one thing they wanted to be mad about — even if it barely related to the app — and they were like: This is my chance, I’m going to do it.

You came away from a hearing that was more confusing than comprehensive, which is why I was thinking I’m like: Everyone else, rightfully, is going to be pointing out how batshit this entire hearing was. They’re gonna do a lot more eloquently than I would. Also, I’ve been writing that tape for four years. So I was like, cigarette in hand: What if I take these people at their word, and try to see if there’s anything salvageable in here? I kind of noodle around with it a little bit. I literally made a pie chart when I was listening to the hearing, and every time I heard a claim that was actually pretty grounded in reality, and then immediately heard something that was barely coherent. I was like: Okay, maybe we can make like some sort of like fractions out of this.

In my newsletter, I was like: Okay, if you go through every single claim— which I didn’t, I went through the first hour — after that first hour, what I came away with was maybe seventh, if I’m being generous, maybe like a seventh of their claims were sort of legitimate. The rest of them were either wrong, or just absolutely ridiculous. They wouldn’t be addressed by some sort of ban at all. Or they were not only TikTok’s problem, but Meta’s and Twitter’s and Google’s as well. I mean, if the first hour is representative, then there’s kind of your answer.

PM: Absolutely. I think you can see, on one hand the congresspeople tend to use these hearings in that kind of way, where they want to get their position in, so that their supporters can see it and they can take the clip and put it elsewhere. But I think you can also see that this is a CEO from an app that operates in China, that’s a Chinese company. That is not one of the big American companies that, they still kind of do want to do well, because it’s good for the American economy. In this case, they really don’t care and they don’t mind just really taking out all of their anger about social media more broadly, on this guy who’s come over to answer their questions.

SW: During the hearing, a lot of these senators were championing their own version of a privacy or a tech bill. I heard them basically doing these promotional laps around the CEO. And I’m like: You want to look hard on Big Tech to get support for this bill that you have, and it just so happened that this guy was in the hot seat. That’s why I said came away feeling bad, because it really was a wrong place wrong time. Obviously, any multibillion dollar conglomerate does not need our pity. But I was looking at him, and I’m just like: Aw, I have a bleeding heart about this stuff. It’s my fault.

PM: I do want to get into some of those specifics that you were talking about, some of the claims that they were making. I pulled out some of these things, as I was reading your piece that you wrote, which, of course I’ll link to in the show notes. I was reading one quote by a Republican Congressperson, Cathy McMorris Rodgers, and she says, “The Chinese Communist Party can use this as a tool to manipulate America as a whole.” I feel like that plays into this broader assumption, or this broader accusation, that’s being made around TikTok that not only can it collect a ton of data, but it’s funneling that data back to the Chinese Communist Party, which is getting all this stuff on the American public, and using that to shape what the American public sees, to engage in these misinformation, disinformation campaigns that we often hear talked about. I wonder how accurate these arguments are around TikTok’s data collection, handing that to the Chinese government, and all these things that are associated with it. What do you make of that?

SW: I am a little bit rusty because it’s been a while since I’ve had to make a living off of following this stuff. But there’s been a lot of really good substantial research, Citizen Lab, in particular, has done a lot of good stuff here. They break down exactly what the app is collecting and where it’s going. TikTok, the US company, which is legally separate, I believe, from Douyin, which is back in mainland China, works off of servers believe in California.

PM: They’ve also made a deal with Oracle, that you reported on a few years ago, where a lot of their data from the United States is stored on their servers, right?

SW: Exactly. But when you talk about data going from the phone that you’re holding in your hand onto a server in San Francisco, or in China or wherever, it’s not always a straight line. In my writing — because I write about advertising a lot, like advertising tech, typically not on platforms is largest TikTok, but sometimes that changes — there will be the sorts of intermediaries that can swipe in and siphon off different pieces of data from the stream that’s going from your phone to that server. Sometimes it stops by multiple servers, and so I believe, if memory serves, we have no evidence, when I checked this two years ago, I believe, I found American servers being pinged and nothing else. We have nothing on paper that shows data from the apps in an American’s hand, going to a server back in Beijing.

That said, there’s some Chinese national security laws that basically mandate, if a government official there asks for data from TikTok, they could say no and kind of deny it. But in the past, when tech companies have had foreign workers operating in Beijing or Shanghai, or anywhere around there, there’s stories that have come out about the Chinese government making these employees life, basically, hell until the American company kowtows to their demands. When I was covering Trump’s ban, three years ago, it was a different world back then. Donald Trump was President, we were all stuck in our homes, everyone was freaking out. I believe you felt this, this time around with your own piece, I was getting a lot of pushback that my coverage was ignoring the “Red Threat” that China could bring on all of us.

I was like: You know what, a lot of those critiques were in really bad faith and just ignored what I have written. But the same way that I took lawmakers at their word this time I was like: Okay, I will look into this and see how other tech companies have fared. And it turns out, they haven’t necessarily fared that great. The Chinese government can be a little scary. So instead, my question wasn’t: Will a TikTok ban do anything of substance? Instead, my question was: Let’s assume TikTok gets banned, why not? What else could the Chinese government do? What other sorts of methods do they have at their disposal to get into all of our pockets? It turns out, I found a lot, circumstances depending.

PM: I think you put it really well there because this is a really important angle on this story. The thing that we need to understand is that the way it’s often being presented by people in government, is that if we ban TikTok, then that cuts off the ability for China to access all of our data because we’re using this app. But if we actually look at the connections, as you’ve outlined in some of the work that you’ve done that exist between all of the apps that we use on our phones. Whether they are ones that are created by Chinese companies, or not.

Even social media platforms, like Facebook or Twitter or Instagram, that are controlled by American companies, that we’re all very familiar with using, because, as you were saying, of the advertising that happens there — and the data that’s collected for that advertising, by intermediaries and data brokers — that because of that, that offers many different ways that that data can get back to China or to anywhere else, or to any other actor who might want to buy it or get access to it. There’s been reporting that at the FBI, that the Department of Homeland Security is buying data from these data brokers as well. That shows us that just shutting down TikTok is not going to be a way to solve this problem, if it’s something that is even happening in the first place.

SW: There was actually a story that came out pretty recently that was really apropos to everything that’s going on right now. Do you remember way, way back when the gay dating app, Grindr — everyone knows it; it’s everyone’s favorite — I’m not sure if you remember, but Grindr was actually owned by a Chinese gaming company for many years. There was some sort of legal situation where they were goaded into selling it to an American group. Smash cut to, I believe it was earlier this year, The Washington Post put out a story where this priest was outed by some shady group that had somehow gotten his location data using a gay dating app that he had secretly hidden on his phone. So this shady group had full carte blanche access to his location. Grindr denied that it was possible, everyone denied that it was possible. And it was owned by an American company, so what’s the big deal?

The methodology that this shady group used to get that priest’s location and track him from his home to the gay club that he wasn’t supposed to be at, and then back home. The way that they did that was, there’s a data flow between that person’s phone and the servers that the company owns. When it comes to advertising, you have sometimes 10, or 15 or literally hundreds, of other folks in the middle, that all operate certain things. Some of them will be doing analytics, some of them will be doing user acquisition, some of them will be serving ads or measuring how well those ads perform. And what this shady group did was, without Grindr know-how, they were able to set up shop somewhere between one of those, possibly hundreds of middlemen, and blend in. Because of the way data works, they were specifically able to siphon off location data from that stream.

So if this stuff was used to out one guy, but that’s not the only group that’s doing something like this. It’s kind of an open secret in the ad tech world. Apps or websites have an agreement with advertisers and maybe a few middlemen, and they shake on it. But occasionally, a bad actor in the US, or otherwise, is free to swoop in and snarf stuff up. In fact, you’d mentioned a piece that I’d written three years ago, where some of those cross border partnerships are actually codified. I’ve written about the company formerly known as Facebook, they have an active program where they work with Chinese advertisers. I know for a fact that there’s ad agencies in Beijing that regularly work with Twitter and Snapchat.

Because of the way advertising works, you need data from a person’s device in order to run that ad in the first place. Sometimes it’s location data, sometimes it’s an IP address, but something’s being traded off there. My pet peeve with a lot of this stuff, is when people talk about data protection, they never really clarify what kind of data. Which it sounds like I’m being kind of an asshole, and yeah, I am, but at the same time, that shady group was able to take that priest’s location, they didn’t know his home address — they had a geofence, like some sort of a rough spot on the map where he was — they weren’t able to get his name, they managed to figure out who he was because they saw him go to a house, and then to a gay club. Then they’re like: Oh, that’s this guy’s address.

So are you worried about your precise location getting leaked? Sure, of course, a lot of people are. There was that story with Strava, a few years ago, where military people were having their locations leaked by the app. So you can take care of location and iOS devices, thankfully, do this. But what about something like your credit card history, or your work history, or who your fiancee is. I you talk about data, like it’s one sort of a amorphous solid lump, the way that congresspeople have done until now, you end up with privacy legislation that’s full of holes, and really easy to manipulate in big tech’s favor. Which is what we saw in California, and why they needed to update their law, the CCPA, after a year.

PM: I think it’s a really good point and it gives us important insight into how this actually works, and why a TikTok ban doesn’t address the problems that these legislators claim that they are interested in addressing. Because if you look at it, if you’re just concerned about Chinese apps, there’s many other Chinese apps that people are downloading from the App Store, not just TikTok. If you’re actually concerned about all these data issues, these are issues with many other apps and not just TikTok.

SW: No and I mentioned in my newsletter, as of that writing with the top 10 apps, two of them were from TikTok, but two of them were Chinese E-commerce apps. And by default, an E-commerce app is going to have your home address because that’s how it’s going to mail you those shoes. I was like looking at it snd I’m like: It it because TikTok is presumably getting that data and manipulating us? Other apps do that too, there’s countless stories of foreign regimes taking advantage of Twitter mobs, for example. That’s besides the fact that Twitter already is giving data to folks abroad, that are also beholden to the same laws that the Chinese government has in place. So by banning TikTok, you’re just gonna make a lot of teenagers mad, more than anything. Obviously, despots and bad actors, they’re not going to go away just because TikTok, isn’t there. They’re just going to follow you to where the people that’s how they work.

PM: Absolutely. To pull another quote from the article that you wrote. This Congressperson that I mentioned before, Cathy McMorris Rodgers, she said in this hearing: TikTok has repeatedly chosen the path for more control, more surveillance and more manipulation. As if every social media app hasn’t done the exact same thing, but you mentioned the algorithm there and the idea that TikTok is making us believe certain things and whatnot. What do we know about TikTok’s algorithm and what they’re actually doing and how this works? Because this is something that the American social media companies seem particularly concerned about because the TikTok algorithm does seem to be more effective than what they’ve been doing for a number of years now.

SW: To back up really quickly. Over the years, because I started covering TikTok privacy specifically way before it was this blockbuster app. So for a long time, I got used to following developers, based in China or based in Malaysia that were working for the company, there just wasn’t anybody in the States back then. I also noticed, they tend to be more open about sharing exactly what’s going on inside the company there than they are with folks in the US. I believe one of the lawmakers during the hearing asked: Why are you so nervous? Why are you trying to bury your company’s Chinese ties. I’m just like: Gosh, I wonder why. Because then this shit happens. While I was writing up this three thousand word newsletter, I wanted to get the Beijing perspective on what was going on over here and I noticed a lot of folks linking to a specific kind of set of blog posts and like developer documents

I should note, a lot of the stuff was written in code, a lot of the stuff was written in Mandarin code, specifically, with a lot of jargon that doesn’t translate across borders, but I did my best to decipher it. What I found was either enough ByteDance folks had spilled the beans, ByteDance being TikTok’s parent company, or enough people living abroad had successfully reverse engineered the app. That they had essentially broken down what makes a lot of this stuff, so addictive. I have about like, 500 tabs with the stuff literally, so I’ll put most of it in the newsletter. But the long and short is American platforms, like your YouTubes, your Instagrams, that sort of thing. They tend to throw the most popular person that you might be interested in. Like I mentioned earlier, I watch a lot of sewing videos because I’m trying to learn how to sew.

So on Instagram, for example, when I get recommended new people to follow, it’s dressmakers that have thousands and thousands of followers. Because the thought process there is, if five thousand people already liked this dressmaker and you like dressmakers, chances are you’re going to like her too. It makes sense. TikTok, at least according to what these people were saying, basically takes that formula and throws it out the window. TikTok does not really give a shit about popularity, and instead, it will surface random clusters of stuff that’s related to what you’re into. It will just throw it your way, and if you like it, you like it.

If you don’t, you don’t, but if you do, and if enough people like a certain video clip within a certain timeframe, it gets recommended to a small circle of people at first. If it reaches that threshold gets recommended to a larger circle and then a larger circle and then a larger circle. But the way that these people were saying it’s set up, is that everyone has a fair shot, which is why, on Instagram, I was being told to follow dressmakers that have these really pretty aesthetic flowy wedding dresses, but on TikTok I got I recommended a person with ten followers who makes furry suits. Because TikTok, was just like: I don’t know, like, I guess you’re into this now, and I totally was. TikTok, will never say this stuff out loud, so basically a lot of this is hearsay. I don’t know, it makes sense to me.

PM: This is all the proprietary algorithm, so they’re certainly not going to share it with us.

SW: Like, you hear me say that and it all sounds pretty benign. I’m talking about wedding dresses and fursuits. This hearing just made me more confused than really anything. I mean, I was angry, but I was also really confused. I was wondering, I’m like: What sort of media do they think a foreign adversary is going to push in front of us? Which, again, that sounds like a really like: Why do you care about this really specific thing? Like you’re ignoring the big picture. Which like, yeah, maybe I am. But if we don’t know what we’re scared of, and how that manipulation would happen, then you end up with accusations that are either wrong or could apply to literally anybody else.

Also, I’ll just say this, before getting into all this stuff. I got a neuroscience degree, and I focused on learned behavior specifically. A lot of sort of the manipulation that’s being alleged, not only of TikTok, this is a platform thing in general, a lot of it’s alleging nefarious intent on their part. But really, people are just dumb, you see a pretty picture and you click on it. My sort of takeaway from a lot of this stuff, if you want not to be manipulated, you have to put in the legwork to try not to be. I don’t think a lot of people know how to do that, so instead of asking, they’re just pointing the finger somewhere else.

PM: I think that in having you described that, though, that also points us in the direction of: If we were really concerned about these problems of data collection, of the way that the algorithm is used. If some of the other things that these lawmakers were pointing out when they actually made sense and weren’t just completely ridiculous. That if we really wanted to address these problems, it’s not a TikTok ban that you need to do in order to address it. It’s federal data privacy regulations, regulations on algorithms, and other things like that. But these lawmakers do not seem to be proposing those sorts of solutions. So what does that tell you about the actual objective of this campaign if it’s identifying all these problems, but then not really offering a real solution to them?

SW: No, during the hearing, a lot of folks brought up that this is an example of why we need a federal data privacy law. And I would be chuffed to bits, if this nonsense ended up being the reason that we finally get that law because right now in the US, different states have different rules for what you can do. As a result in my interviews with sort of these middlemen and shady characters, instead of abiding by one law in Virginia and one law in Connecticut, they just ignore both of them. And because lawmakers on the grounds in these states are so busy holding hearings that go on for five hours, they don’t get caught. Really the only place that has any sort of substantial ground here is California with The CPRA. But again, they managed to get got so good that they pass the wrong version of a law the first time around.

PM: I think they’re all good points. I would say that, I’m obviously quite pessimistic on everything that’s being proposed here. I think that really at the core of this is not really a desire to address these fundamental problems with social media and all of these apps that we’re using constantly. But to take on one of the competitors. To the major social media companies that exist in the United States, instead of addressing the core of all their business models as well.

SW: I mean, I’ve covered the company formerly known as Facebook quite a lot in my time. I noticed that — this was like around 2020, during the Trump administration —Mark Zuckerberg would very often position his app, successfully, as the savior for small businesses, because it is true that many small businesses, by default, are dependent on Facebook’s platform to reach their audiences. That’s true. Everytime, a very reasonable critique was aimed at the company, they would just immediately pivot to: If you regulate us, think of all the small businesses that would be harmed. Which yeah, is true, but you can’t have it both ways. I thought it was really funny from the day of the hearing, that TikTok flew in all of these influencers, who also have their livelihoods depend on that app.

I’m like: Oh, we’re just using literal human shields now. And it has worked for Facebook in the past, because nobody wants to be the guy that toggles your local ice cream store just to get back at some company in Silicon Valley. So in this case, if I’m thinking about most of the American public, I feel like going with the small business angle might be a better sound to Congressman’s ears than influencers. When really, on the ground, they are the same thing. Luckily, for TikTok, it’s been leaning really hard into E-commerce lately without much success. But if you read the tea leaves, you can see the company priming itself to be like: No small businesses depend on us too. Which, again, I have no sympathy for multibillion dollar companies. But I’m watching it, and I’m like: We didn’t learn the first time and I don’t know if we’re gonna learn now.

PM: A perfect point, and I would just say that Amazon does the same thing. I was talking to Moira Weigel a few weeks ago. They basically say: You know, we’re protecting all these small businesses, making it possible for them to exist. There’s something very pernicious and self-serving that’s happening there as well. So, Shoshana, to close off our conversation, we’ve been talking about several different aspects of this. Is there anything else that you think people should know about this hearing in particular, or about this wider discourse around TikTok, that we haven’t gotten to in this conversation?

SW: Well, if they want to stomach my more than three thousand word newsletter, and I’m gonna keep writing all this stuff, so you can definitely follow that. Quick little plug there. But more than anything, when you read news stories about this, or news stories about any app, really, and you hear claims that seem to be legitimate and seem to have backing and sound true. Just always ask: What exactly is this person afraid of and what am I afraid of? Does this claim address that? Because you’ll typically find that it doesn’t.

PM: I think that’s always great advice for people to keep in mind when they’re reading these stories and considering the narratives that we’re getting around these tech companies. Shoshana always great to have you on the show. I really appreciate you taking the time and I can’t wait to have you back on again. Thanks so much.

SW: Aw, thanks!

Paris Marx: Dan, welcome back to Tech Won’t Save Us.

Daniel Greene: Thank you so much, Paris, happy to be here.

PM: Very excited to chat with you, again. It’s always great to have you on the show and to get your insights on the tech industry. Of course, in this moment where we’re seeing the US government’s policy on technology and the internet seeming to shift in response to, I would say a number of factors, but in particular, the competition from China. I think it’s a good moment for us to revisit a bit of history to inform a conversation around what’s happening now. I want to start with a bit of that historical piece just to inform what we’re going to talk about. I think, it’s important to understand where this framework for the US government’s approach to tech and internet policy really came from. So what was the orientation in the late 1980s through the 1990s, as the US government was formulating its policy on internet and technology?

DG: As with everything else in American economic history, it’s history of blowback, about overreaching in one particular area and then that coming back to bite you in the butt a couple of decades later. So the commercialization of the internet really cannot be divorced from the moment of globalization that follows the economic crash of 1970s. The nationally-oriented Keynesian economies of 1970s start to fall apart for, pick your poison, for Gold Standard Oil Crisis, drop in productivity, whatever. And we see then a move to much more nimble supply-chain., stretching all over the world. Especially start to encircle countries which maybe had tried to create their own welfare states, but then were ruthlessly gutted by the Volcker shock by debt imposed by the IMF, etc.

At the same moment, we also see that the Democrats have been really out of power for a long time in the US with the semi-exception of Jimmy Carter. During the 80s, during this time in the wilderness, the American Democratic Party is trying to figure out how to get back in and they do so in part by proposing the internet, and the class of people with an interest in the internet, as the solution to the problems created by deindustrialization at home. The late 70s and 80s are also a period of heavy deindustrialization, although that’s a process going back as far as immediately after World War II. So what Clinton and Gore and other folks in this Democratic Leadership Council do is both bring in a new class of voters and activists, very different from the tense but multiracial working class coalition that undergirded the New Deal.

We have a lot more professionals, people who work in office parks, people who are very committed to formal equality, but not necessarily equality of outcomes. So, against housing discrimination, very a protest against that, not so much fan of public housing being built in their neighborhood. These folks often end up in office parks producing high technology products, Montgomery County, Maryland, Route 128 in Boston, Silicon Valley. These are what are called the Atari Democrats. A name that they end up giving themselves. That comes along with a new group of investors at the top of the party, including some people like the heads of HP and Apple at the time, who were lifelong Republicans. They were a little concerned about the isolationism that Reagan promised and the threatened defunding of places like the Department of Education, because these were industries that were extremely capital intensive, that had really broad ties to the global economy.

But were also producing cultural products, things like software, or media. Things that depended on English remaining the language of global commerce. So this new group of financiers — of tech executives of capital intensive manufacturing, distinguished from especially the oil people that really supported Reagan — started to reframe the party, at the same moment that NSF Net was being transitioned towards privatization as the internet. So by the time Clinton wins the election in ‘92, they are proposing that the internet is going to be the thing that is going to take care of the industrialized US and secure American hegemony abroad. They have just defeated the Soviets, they’re still a little scared of competition from Japan and West Germany.

We have the most robust internet infrastructure here in the US at the time. Extending that globally and building up abroad is a way to connect everybody to high-pain, knowledge-work jobs that do not respect geography, unlike those old fuddy duddy manufacturing jobs. So that promise to solve the problem of deindustrialization comes out at the exact same time as the people who would most benefit from internet liberalization, have taken control of the party. This is how they end up distinguishing themselves from Republicans and this is why the 1990s are the height of American liberalism, it’s the thing that everyone keeps trying to return to, because that was the moment when everything was going really right for them. What we’ll see is that a lot of the decisions they made on things like liberalizing telecommunications infrastructure as a condition of entry into the World Trade Organization, that then comes back to bite us in the butt. As, like Huawei, ends up taking all of the infrastructure projects for telecommunications across the world.

PM: It’s such a fascinating history, and you describe it so well there. I would be remiss if I didn’t say, if people do want to know more about what’s going on in that period, of course, we talked about it more last time you were on the show. Of course, they can go back and listen to that interview too. But now, based on what you were talking there, obviously we see that there’s this move toward the internet and privatizing the internet, liberalizing the internet, having the internet go global. Some of the narratives that we usually have around this period is that because the internet is being liberalized, becoming more accessible, the open Internet is expanding everywhere.

There’s this great expansion in freedom of speech and freedom of expression that is enabled by that. And I think that it can be easy to see that period, solely or primarily, through that lens and not focus on the economic side of what was actually going on here. So you talked a bit about it there, but what was really kind of the goal of the Clinton administration and Vice President Al Gore, as they were pushing for this internet to be privatized, to be commercialized and to be extended around the world so that many other countries were using this network that was originating in the United States.

DG: The internet is a network of networks that not any one person can claim ownership to. What we’re really seeing with American economic and foreign policy at this moment is a way of identifying a particular setup of relatively frictionless communication between states, undergirded by Western, but especially American infrastructure, and Western, but especially American software firms, and networking firms. So Microsoft, Cisco, that being the engine of liberalization, and there are plenty of people in the US and out who have, perhaps, libertarian visions of an open internet, but one that is agnostic on or even opposed to American interests. The more anarcho-punk approaches to cyberspace.

But the Americans were quite clear on both economic interests of the US and liberalization of the internet being the same thing. If it’s okay, I’d like to read just a really brief thing that Al Gore wrote in 1991. So there’s this special issue of Scientific American with all the heavy hitters of tech libertarianism— Negroponte and those sorts of people— and Gore was the champion of liberalized internet in the Senate. He then took it on as Vice President, I do not joke when I say that I really do think that Al Gore created the internet because his vision of what was going to happen is the world that we live in now. He saw himself very much as the 1990s equivalent of his father, who was very, very integral to the creation of the American highway system. Which was itself a Cold War project to make us cities less reliant on single nodes, and to increase the flow of goods between cities.

With that in mind, Gore in Scientific American in ‘91 writes, quote: The unique way in which the US deals with information has been the real key to our success. Capitalism and representative democracy rely on the freedom of the individual. So these systems operate in a manner similar to the principle behind massively parallel computers. These computers process data, not only in one central unit, but rather in tiny, less powerful units. Capitalism works on the same principle, people who are free to buy and sell products or services according to their individual calculations of the costs and benefits of each choice, process a relatively limited amount of information, but do it quickly. When millions of individuals process information simultaneously, the aggregate result is incredibly accurate and efficient decisions. Communism, by contrast, attempted to bring all the information to a large and powerful central processor, which collapsed when it was overwhelmed by ever more complex information.

So Gore is very deliberately saying here that the liberalization of the internet and extension of that openness to the rest of the world is identical with American capitalist interests. They are the same thing and they will reward everybody else, but it’s still his rules and his country’s tech, at least in the beginning, that is undergirding that project, they are the same thing for them. In terms of how it ends up happening, there is some argument that American contractors and the American state should shift some money from defense spending, no longer needed as much for the Cold War, into internet infrastructure build out. That’s more of a campaign tax than anything else, the actual money distributed, never tops a couple of billion dollars, it’s really nothing compared to the Pentagon. It doesn’t even really compare to the Crime Bill in ‘94.

But it was important rhetorically, and it was important to build the identity of this new political coalition. To the point that when Gore is back on the campaign trail in ‘96, he’s talking about his opposition and the Dole campaign and saying: These people that want to defund our internet build out they’re unilaterally disarming us against international competition. So it is very clear here that they see international economic liberalization, and American interests and the growing internet, as the same thing. They’re all identical. The problem that we come into a couple of years later, a couple decades later, is that those things are no longer identical, in part because that hegemonic coalition that Gore and Clinton were constructing around the internet begins to fracture.

PM: I think really well put and I think it’s so important to understand Gore’s orientation toward the internet, and how that really sets up the internet that we have today.

DG: To put a fine point on it, he sounds like [Friedrich] Hayek in that. I mean, the argument that was made 70-80 years prior about why communism couldn’t work is because it was an inefficient economic process. Then you have the leader of American Liberalism, saying the exact same thing for his pitch as a New Keynesian. This is why we need to invest in this thing, because it’s American capitalism. It’s a relatively coherent narrative from the free market libertarianism of even the 30s, up to the internet liberalization that we start to see in the 90s.

PM: A great point and a key one. I have one question that maybe starts to bridge this period with what we’re looking at now. And this is to say that as you were talking about this idea of the open internet is coming along at a moment where there’s a lot of neoliberal policies that are emerging. There’s the effort toward globalization, and the US is creating a global system that is oriented around ensuring that Western capital can easily access more markets, so capitalism continues to expand, and all these sorts of things. I feel like when we look at the orientation toward the internet, or if you look at the internet from a US perspective, it can maybe not seem so bad. That US companies are going abroad, and that everything is open and whatnot. But I think that when you look at it from a foreign perspective, you can very much see how, as the internet goes global, it ensures and it allows us companies to really take over the sectors that are associated with the internet, and that the internet is expanding into.

I think that for a while, it’s difficult to see much pushback to that, or at least not pushback that gets a lot of attention. But in recent years, we’ve begun to see more and more of that, with Europe getting more frustrated that so many major US companies dominate its’ tech sector. Obviously, with China and its ability to build competitors to that. Other countries as well had been speaking out more about how these US companies dominate the technology industry in their countries and make it difficult for their domestic competitors to compete, if they are not allowed to take measures that restrict the ability for us tech companies to dominate. And so I wonder what you make of that relative ignorance of that aspect of US tech policy, or the ability of these US tech companies to really dominate internationally in that way, and how that is really begun to shift in recent years?

DG: I mean, at least in the 90s and early 2000s, I think, frankly, it’s the success of American imperialism. So the Soviet Union is defeated and the US is the representative of global capitalism at that point. And I pretty much agree with Sam Gibbons analysis here that like capitalism is global, but it is protected, set up, restored, spread through American imperialism. Especially in the 90s, that’s the only game in town. Chinese military is too weak and the workforce is too far down the value chain, ditto for Latin America and Africa. Europe, at least at that point, is mostly happy to identify its neoliberal ambitions with that of the US. So engagement with the WTO, or a general agreement on trade, proceeds largely on American terms. That begins to fracture over time. The earliest protests that you can see against that are largely identified with the same kind of content protectionism that we saw earlier in the century.

So the French effort to create Minitel is super interesting and is largely building off the French content industry that is protected and promoted, and subsidized as an effort to protect French culture. And it’s not that different from like Canadian or Québécois, content being produced as protectionism. Minitel is probably the best national example of something that was built to oppose the American internet. It did work very differently, but it lost, it flamed out. I think, especially in the 90s, and this is in part why American liberals are so nostalgic for that period, it is the strength of American imperialism and the ability to enforce those trade agreements to global scale that allows Americans to think: Oh, yeah, my interests are totally identified with the rest of the world’s interests. That’s the definition of hegemon, when you’ve imposed your particular class interests onto everyone else as, supposedly, a universal interest.

PM: It’s a really important point to make and an important thing to understand about how this policy ultimately works, and why there’s so much excitement about the global expansion of the internet, even though there are other things that are associated with it as well. Obviously, there’s a lot that happens in that intervening period, we’ve talked about pieces of it and we certainly can’t get to everything. But in the past number of years, we have seen the beginnings of a more significant shift in the US approach to its policy on the internet, internet infrastructure, and other aspects of the tech industry, that we haven’t really seen as much, I would say in those past recent decades. Where we have seen actions under the Trump administration around companies like Huawei, and now these things continuing into the Biden administration, with the CHIPS Act and the potential action against TikTok, and other things like that. What explains why the United States is making these radical shifts away from this more open approach to technology and the internet in this moment?

DG: It’s probably most helpful to talk about things that are not specific to the US. I would frame it as two general trends, and one is what the tech sector — I think we can say that quite narrowly in the 2000s, 2010s, even the 90s, identified with software development — and what software development did to national economic developmenta and a bunch of different places, including China. Then how the rest of the international political economy has changed since the height of free trade liberalization in the 1980s. So I think it’s just a reality that by the mini-recession that we get during China’s withdrawal in the 2015-2016 era, which comes back to hit especially American Rust Belt states pretty hard, may have helped get Trump elected.

We have a growing dissatisfaction with the tech sector, what we’ve often called the techlash. This, I think, accelerates after Trump’s election in the US, especially because of the concern of disinformation. This is a broadly shared sentiment across the entire world. Part of it is political, folks, everywhere, are extremely worried about social media, and the power that a few American companies have to determine what counts is speech, and large swathes of the world. In some places, especially Facebook is identified not just with the content provider, but the entire internet through the way they package themselves with mobile services. So a lot of people were worried about the power of these platform companies and their effects on the macro-economy.

It became pretty obvious to a lot of people — and I think this is the driving force behind what people have called Biden’s Supply-Side liberalism — is that software development made a lot of people very, very rich. There was even less trickle down there than in other places where you expected some trickle down, in part because of the production process. You don’t really need to employ that many people. They’re all of a relatively similar background. And the externalities are not that serious in the way that you would think of large manufacturing and or even a big school or something like that. They’re not building up larger ecosystems of anything besides software developers. People talk about the Japan-ification, of industrial economies in the post-Recession period as being a lost decade. We really did not have much wage growth, or much job growth, post-Recession, especially in Europe.

I think there is this growing resentment across the world that placing all our eggs in the American software development basket just did not work. It just did not give us the job growth, it did not give us the wage returns, it did not give us the things that we needed besides people being priced out of major downtown cores. So that’s one thing, everybody is feeling that all at the same time. So the US is getting more suspicious about software developments, about these large platforms, at the same time as everybody else’s. You can see that in China, with the recent crackdown on tech executives, which is part of a larger crackdown on things like these tutoring companies and stuff like that. But there is a sense that these people have gotten too big for their britches, and have distorted the local state, that’s interpreted in China as they’re challenging the party power. But same thing elsewhere.

The other thing that’s happening is that the liberalization seems like a great idea for those people that are already on top of the value chain, and are saying: Okay, we’ll privatize all of your country’s resources and maximize its openness to the free market, that’s going to be great for us because we have a bunch of new low wage exporters to process our iPhones or whatever. That was really like the American Dream for trade liberalization is to see China as the world’s factory, that was great. They really liked that. The problem is that China in particular, I would say —absolutely South Korea, Japan, and West Germany, and Vietnam to an extent — these places did what everyone said they should do, which is move up the value chain throughout that period of globalization. They often did it in ways that were not the American path to doing things, especially in Vietnam and China. You have a fairly strong central state that is orienting this economic development.

But these places start to move into value added manufacturing, they start to move into infrastructure development, they start to move into services, they start to move into web development. I think, frankly, a lot of the American reaction to TikTok in particular is: Wait a second, I thought we were the software guys, I thought we were the platform guys. I thought we were the surveillance capitalism, guys. Now, you have some Chinese guy coming in and telling me what to do with that?No way Jose, that’s my business. So I think what they had hoped for in the 90s was a global market for American goods and global suppliers of low wage labor to put those goods together. But what actually happened is that especially Chinese companies end up competing with a leaner and meaner capitalism that Americans did.

I remember, in 2015-2016, when you had Chinese venture capitalists in a room, they would usually say, and not incorrectly: If you want to start a company that can run into regulatory problems — something like computer vision, or something, which is always a touchy surveillance topic — you should come to China, you should do it in China, you will have less resistance from the state here than you will in the US. We don’t have all those burdensome regulations, because I can pay off my local party guy. That may have been true for a while now, I don’t think it’s true anymore, but the openness of capitalism that was pursued in the 90s ended up all through the stack — from the chips to the infrastructure to the web development — ended up producing a new set of lean mean companies that really challenged American hegemony, and Americans are not used to that.

PM: When we see these developments it’s very clear how the Chinese have become very competitive with the United States, and that this is driving, in particular, much of the response against Chinese companies, because they’re the biggest competitors that US firms are facing. I wonder, you talk there about how the United States saw itself as making this software and designing these things. Then China in particular, was going to be the factory where all of this was created. Then that was how this arrangement was going to work. But it seems that especially when we look at Huawei — and to a lesser degree the CHIPS Act, and some of these other concerns about the critical minerals, and electric cars more broadly, because this is also coming in with the regulations that the Biden administration is putting on electric car development and things like that — it looks like what we’re seeing is, in part, the consequence of that decision. We’re not going to make very much in the United States, we’re going to allow it to be made in China. Then as a result, China is learning from that manufacturing process, moving up the supply chain, gaining additional skills, and coming back to be able to compete with US firms who have lost a lot of that institutional knowledge on the production side, even as they’ve been doing something else.

DG: In general, what you’re seeing with the Biden administration, in this Supply-Side liberalism, is a recognition of the divergence between the interests of global capital, if we can say there is such a thing, and the American state. In the 90s, it was very easy to say those two things were the same and perhaps even through the 2000s. The process whereby pretty much every American chip maker, besides Intel, no longer makes their own chips, and this is relatively recent development. So there’s a guy, Jerry Sanders, who is head of AMD, said at some time in the 90s: Real men have fabs, real men have fabrication plants where they make their own chips. AMD now, does not make its own chips. But this was seen as a post-war style of vertical integration, where you would design the thing and build the thing.

These days, America has a ton of chip designers, they’re at the top of the value chain there. But bar Intel, and some folks lower down on the value chain who make single use stuff for your car windows or whatever, we mostly don’t fabricate a ton of chips. The market is still very concentrated — it’s pretty much just South Korea, the US and China and Taiwan, and China is really a distant fourth — that fabricate chips. But today, the most important chip making company in the world is TSMC, it’s Taiwan Semiconductors, and they do not design their own chips. They are a foundry, who takes other people’s designs and then manufactures them at scale. Probably the most important company in the world, in terms of its strategic value to everything that we make, some people say that 50% of devices with a semiconductor in them have something from TSMC, they produce Apple’s chips, for example.

They’re in Taiwan, they’re right across the street. If stuff pops off, that’s the linchpin of the global economy right there. That’s the moment when folks in the national security apparatus start to get a little nervous and they say: Well, it’s probably not true that China can just take TSMC and start producing its chips just like they did before. For all the reasons you said there’s a lot of like skill involved, a lot of learned things. Especially in chip fabrication, the design and fab work hand in hand. There’s a lot of trial and error as you learn to use the tools and you adjust them to this particular place and it feeds back into the design. But it still makes people really nervous that that kind of thing is right there, right across the coast from China in a country that China does not believe exists. So I think a lot of the Supply-Side liberalism on ships, and then especially with infrastructure in Huawei, which is not really a place that Americans compete with. Most of Huawei is competition is Scandinavian, it’s Ericsson and Nokia.

You’re recognizing that the military is scared that they don’t have control over these crucial components that are essential to everything in the world, but especially to American military superiority. American military superiority and free trade was very visibly linked with the smart bombs that we saw in the 90s, live on CNN. There’s a decent argument that that is, what is giving Ukraine such an advantage over Russia right now, or at least what is allowing Ukraine to prolong a protracted war that you would think that a much larger military would be able to win very quickly. So we have seen that the interests of the American state and global capital are at least intention in this way. The CHIPS Act is designed to correct that, it’s not a ton of money.

I think it’s like something like $40 billion in incentives for fabs and then more for R&D, 40 billion is like one fabrication plant that takes three years to get online, and is not going to be a top of the line thing that you would put on an iPhone. Those are more mature markets for chips that would go inside cars or toys or something like that. But it’s trying to reassure or at least ensure companies like TSMC, it’s trying to invest in R&Ds that American manufacturing can be more closely aligned with our chip designs. It’s certainly not the moonshot thing that would say like: Intel is our national champion, we’re going to pour so much money into intel that they’re basically nationalized. It’s not that, but it is a recognition that there’s a problem and they’re trying to put the cat back in the bag. Whether that’s going to be successful. I don’t know.

The sanctions against Huawei, a little further up the stack and infrastructure, were very, very harsh for a couple of years and it really destroyed Huawei is revenue. But they’re profitable again, they’re down 30% or 40%, from where they were in 2018 or so before the sanctions started hitting, but they’re profitable again. And they seem to have been able to pour enough money into R&D, that they can develop their own chips and stuff that was prohibited from sanctions with US suppliers. So it maybe that this just forces Chinese companies to create their own ecosystem, which would maybe backfire a little. But you are realizing now that this economic liberalization of the 1990s with these tech markets has come back to bite us in the butt. That’s what the Biden administration is trying to solve, and that’s what the Trump administration was trying to solve. I think there’s much more continuity here than we think.

PM: It’s interesting to see that almost bipartisan aspect to it, where Trump was really pushing a lot of these things, but then Biden has very much adopted them. I think going back to the 90s, when we were talking about it wasn’t just Clinton and Gore and the Democrats who were interested in pushing the internet in this way. But Newt Gingrich and the Republicans were right on board with this broader vision that was going along with that. We were talking about more of the tech manufacturing side of it. We were talking about the CHIPS Act and Huawei, but then there has also been this focus on TikTok, and some other kinds of more technology services that are coming out of China as well that are competing with major service companies like Facebook and Instagram, in this case, if we’re thinking about services. But also you look at Shein and other kinds of E-commerce companies that are competing with Amazon, for example.

There’s recent rumblings that now that TikTok has been in the spotlight Shein might be getting in the spotlight next, it might be the next focus. So what do you make about this aspect of it where obviously there are a lot of arguments that are being made around why TikTok needs to be getting this attention. That as I talked to Shoshana Wodinsky about it doesn’t seem to really be accurate. A lot of the problems that are being identified with TikTok are the case for much of social media, but the regulatory action is only against TikTok specifically, or proposed regulatory action. So how do you see the approach on the service side of things as distinct from the Huawei and CHIPS stuff?

DG: Honestly, I do think the approach Tiktok is relatively consistent insofar as like it’s bad when China does it. That’s consistent. I don’t doubt that there is — despite whatever financial maneuvering that makes it a entirely private company with data stored in the US as your excellent Jacobin article pointed out — that I’m sure that there is still some Chinese Communist Party interest in ByteDance writ large. I mean, they would be stupid not to just as the US would be stupid not to maintain interest in Facebook, Twitter, etc. I mean, we’ve seen plenty of even county level American law enforcement getting cooperation from Facebook. So I’m sure the most powerful military in the history of the world is doing the same and the Snowden revelations told us exactly that.

It’s all through the stack. Everything that they’re scared of Huawei doing is exactly what we’ve been able to get AT&T, Verizon to do. Just China does it, so that’s bad. I think that’s relatively consistent there. The other thing I would say, and I think your point of comparison of Shein is a really good one. It reminded of some lessons I’ve gotten from like scholars of intellectual property and race, like Anjali Vats or Larisa Mann that intellectual property — and you really saw this in the 90s, with the WTO agreements — is inherently racialized. There is a global hierarchy set by these trade agreements, that specifies who is a creator, who is a consumer, and who is allowed to do what creation. Even within Chinese finance circles, or even outside investors who invest in China, everyone was kind of shocked at the success of ByteDance and TikTok, just because that kind of cultural phenomenon was not something that people expected to come out of China.

They had invested into as more as a producer of low value commodities, not this mass cultural phenomenon. And to be fair, it is the only global one that has emerged out of that ecosystem. Alipay to an extent, TV, like the Wandering Earth, kind of thing. I think there is a sense from Western viewers that TikTok is exceptional because it’s doing things that the Chinese economy shouldn’t. It has broken the global hierarchy of creatorship, which is generally a color line. I think you can read the accusations against TikTok alongside accusations of copy theft in China as the place where copyright goes to die. The place that we have to watch out for unless they take our movies and pirate our stuff from our brilliant Hollywood people. It feels like pretty much the same thing. You’re able to put a military gloss onto it because of surveillance advertising, but ultimately, it’s rebelling against people who have gotten too far up the value chain.

PM: I guess, the final question to reflect on all of this is: Obviously the US is changing its approach here because it is facing greater competition from China. It’s also facing as you were saying, the greater scrutiny of the tech industry in general. Do you think that its measures, that its policy approaches, are actually going to work in defending American tech hegemony from the competition it’s facing from China? Or does it feel like the cat is out of the bag here, and we’re in for quite a different tech ecosystem, or whatever you want to call, it moving forward?

DG: There’s not a button that Grandpa Joe can push, just that says: Stop neoliberalism on it. I don’t think there’s any one moment that’s going to tell us that we’ve escaped, whatever earlier era is. This trend towards the more fragmented national internet has been going for a while, and we continue into that direction where for reasons of protectionism for reason of surveillance, both of which can have perfectly legit national interests involved. I wouldn’t like them, but that’ll continue. For chips, specifically, the EU just passed their own even smaller bill. That’s when you realize they’re a monetary union and not a fiscal one. The actual dollar amounts I don’t think are significant enough to really move the needle. I mean, like TSMC is putting something like $20 billion a year into D&D alone, and that’s just that one company. Their largest fabrication plants are worth hundreds of billions of dollars, you’re really not changing the price of those items coming into the US.

I don’t think it’s enough to change the needle all by itself. I think there is an ecosystem effect, whereby, Huawei starts insuring, starts doing all of its stuff in China. The US through chips through the IRA, through, increased protectionism, through strategic decisions at Intel, who may actually fabricate less moving forward. All of those things, I think, start working together with Europe. Europe is frankly a little worried about the IRA and the CHIPS Act. They don’t produce much chips-wise there besides chips for German autos. As all of those things taken together, start to move the needle, where I think you really start to see immediate effects are stuff targeting personnel. So this is most visible in holding the chairman of Huawei in jail for multiple years in Canada at the US’s a request for purportedly providing parts to Iran.

That is visible in a bunch of smaller actions, where the sanctions against the Chinese semiconductor plants, or fabs, the sanctions against Huawei are making it more and more difficult. For example, for people have dual citizenship to work in one country or the other, they have to choose and stick with that choice. It’s making it harder for say, like a Dutch company ASML, who makes these lasers that are really important to cutting edge microchips. They’re the only company in the world that can do it. It makes it much harder for them to work with Chinese companies. I think these kinds of sanctions that are mostly done through the US Department of Commerce, by placing these companies on the entity tist, like that stuff is as much of a threat if not more than any American industrial program. So it’s likely that this old fashioned protectionism is just as effective as building up our own domestic supply.

PM: We are really at a moment here where it does seem like there’s an attempt to shift something in the global order around this. So I think it’s so important not just to understand that history of where this all came from, but also the reasons why these things are shifting in this moment. I’m really happy that I was able to get your insight on this, Dan, thanks again for taking the time to chat.

DG: Always a pleasure, Paris. I agree this is really, really important. We don’t know what’s gonna happen next, we really don’t. I think anyone that tells you that they know what’s going to happen next years should not be trusted.