TED Talks Daily - The potential US TikTok ban — and what's at stake | Clay Shirky

Episode Date: January 9, 2025

The clock is ticking on social media giant TikTok, which faces a nationwide ban in the United States unless its parent company, ByteDance, sells it by January 19. Social media theorist Clay Shirky unpacks why the US is trying to ban TikTok, what it means for the app's users and creators and the implications for national security, freedom of speech, US-China relations and more. (This interview, hosted by TED's Whitney Pennington Rodgers, was recorded on January 8, 2025.) Hosted on Acast. See acast.com/privacy for more information.

Mayday, mayday, we've been compromised. The pilot's a hitman. I knew you were gonna be fun. January 24th. Tell me how to fly this thing. Mark Wahlberg. You know what the difference between me and yours? You're gonna die.

Don't shoot him, we need him! Y'all need a pilot? Flight risk. As a Fizz member, you can look forward to free data, big savings on plans, and having your unused data roll over to the following month. Every month.

At Fizz, you always get more for your money. Terms and conditions for our different programs and policies apply. Details at Fizz.ca. You're listening to TED Talks Daily, where we bring you new ideas to spark your curiosity every day. I'm your host, Elise Hue. Today, we're bringing you a special conversation on the potential ban of TikTok in the United States and how it will have an impact on the rest of the world. Social media theorist, Clay Shirkey sits down with Ted's curator, Whitney Pennington Rogers. Hello, and thank you for watching TED Explainsains where we take the biggest headlines of the moment and offer clarity around what it all means and context on why it matters. I'm Whitney Pennington Rogers and I'm your host for this conversation.

So TikTok launched in 2016 and I don't need to tell any of you, especially those of you watching on TikTok, that in less than a decade, the app has exploded to become one of the most popular social media platforms in the world, with what many estimate is more than one billion monthly active users worldwide. But TikTok success is often overshadowed by controversy connected to how it treats data privacy, its role in geopolitical tensions between China and, well, the rest of the world and an algorithm that could support the spread of misinformation. Last April, the US government passed a law that would ban TikTok in America unless its parent company, ByteDance,

sold the app by January 19th of this year. ByteDance sued, and in December, an appeals court rejected their claim. Now, in just two days, the case will be heard by the Supreme Court, and TikTok's fate and the wide-ranging implications of what happens next lie in the balance. To help us make sense of this moment and why it matters to all of us, I'm joined today by author and social media theorist, Clay Shirky. He's the vice

provost for AI and technology and education at New York University and has given several TED and TEDx talks on the cultural impact of the internet and social media. Hi, Clay. Hey, Whitney. How are you? Good. How are you doing? Good. Thanks for the invite. Yeah, thanks for being here with us. Sure. Well, let's start from the most basic place. Why does the U.S. government want to ban TikTok?

Yeah, well, there's really a question of sort of stated rationales and then sort of what may be behind that. The stated rationale is national security, and this has actually happened twice. There was a proposal during Trump's first administration to ban TikTok going into 2020. That did not happen. The Biden administration revived that. And last April, President Biden signed into law

legislation that would force TikTok to either sell or shut down in between nine months and a year. The president had the option to extend the deadline for this purchase or shut down and then opted not to do it, which is why the 19th of January is the deadline date for TikTok. The stated rationale was that TikTok's parent company, Bike Dance, owns only about a fifth of the company, but has a controlling share because of the way the stock is structured. And because Chinese companies,

in concert with the Chinese government, hand over data much more regularly and at much larger scale than than typically happens in the US and so there was a worry about Control of US data held, you know in the numbers you talked about the intro, you know in the US cases something like 180 million 180 million users in the country alone there was a worry about The data being held by held by TikTok and therefore accessible to bite dance or so or so the

theory was all of this is relatively vague, which is to say there's no particular identification of actual harms that have been caused. There was some talk of US military personnel giving away their position by using TikTok. But of course, that can happen on other social media as well. Bike Dance is a Chinese based company, but TikTok is headquartered in Singapore. They say that no data is held on Chinese,

no American data is held on Chinese soil. And there hasn't been frankly, a lot of technical back and forth about where the data actually resides. The concern seems both broader and vaguer than that. And then when you start to look at other places where those kinds of risks might appear, you don't see the same concern by the US government.

So if you take the idea that TikTok holds data on US citizens and is not itself a US company and therefore Create some risk If you look at TikTok on people's phones and then you look at Temu the Chinese shopping service Which has got about the same number of users in the US as TikTok Temu has much of the same information TikTok has plus they have your credit card and home address who has much of the same information TikTok has, plus they have your credit card and home address. So if you were worried about American information and national security by foreign actors, it's not clear that TikTok is the place you'd start. It is however, highly

symbolically important. And that's clearly part of this conversation, which is that this is a Chinese media company operating very successfully in an American context. And given the decoupling of the Chinese and American, both markets, but also cultures to some degree, TikTok is a, it's a highly symbolic target and probably the symbolism matters more than the actual risk. Okay. Well, there's so much to dive in there into there. And I think one of the big

questions here is what does this actually mean for users of the app? So that TikTok would no longer be available to users in the United States. Come January 19. If you do use the app, there would be no software updates available to you. So eventually it would become obsolete. And this option of selling is something that ByteDance, of course, has pushed back on. Can you talk about some of the reasons why that's not a viable option for ByteDance or something that's appealing to them? Sure. And again, and in part because of the symbolic nature of these conversations, there's ByteDance's stated rationale, and then there is everything else that might go into that same calculation. So, ByteDance's stated rationale is the TikTok algorithm is a kind of secret sauce that they don't want to disclose, and they would

be forced to disclose it if an American company were to acquire TikTok and operate it separately from the ByteDance ownership. There is the TikTok algorithm is important. Essentially what TikTok discovered and then turned into a very successful service. Is that video is special with regards to viral spread. And then if you pay more attention to how users actually watch video, how what do they click on, how long do they watch, when do they click away, you can actually pay less attention to what their friends like. So TikTok's amplification of video is much more about personal interaction with content, with content, than it is about viral spread in the manner we got used to with about viral spread in the manner we got used to with Facebook in particular where your friends would effectively amplify content from elsewhere and would amplify your content to elsewhere.

TikTok doesn't have that same kind of social amplification. That was indeed an important discovery. That having been said, the TikTok algorithm is not so special that there's any kind of secret sauce there. There was actually a document leaked in 2021. Ben Smith, then the media columnist for the New York Times, got hold of it. And when it discovered that TikTok had done a particularly good job of balancing these various factors of when a user would be interested in a certain kind of video, but that there was no signal in there that they were

taking advantage of that other companies also didn't have access to. So the idea that there's some, you know, the equivalent of the formula for Coca Cola for TikTok that they would be forced to disclose isn't really a plausible reason not to sell. There are however, some other plausible reasons not to sell them. The most important of which is if the US forces a Chinese company to divest from a company that is succeeding in the US market, that becomes a template for other kinds of work and for other kinds of essentially attacks on Chinese commercial operations that are outside of China. You can see this in the way the Canadians handle this.

Canadians have also banned TikTok, but they banned TikTok Canada, which is to say they have gone after the company doing business on Canadian soil, but are not proposing banning the TikTok app. So they're more direct about their concerns with the company rather than with the app itself. ByteDance on the other side of that, the Chinese company doesn't want a

world where essentially Chinese media properties can be nationalized through this kind of law, and then forced to sell, partly because they don't want to lose control of Chinese firms like that, but also partly because there's a great concern under Xi Jinping, under the current Chinese administration, that commercially successful Chinese operations not be convertible to dollars on the part of the inventors. They don't want global entrepreneurs from China to be able to re-situate themselves in other countries. They want to maintain a pretty tight relationship with their most successful

business people, and they want those business people to understand that they are effectively in business with China, the government. And so having an event in which some tens or possibly hundreds of billions of dollars was delivered by TikTok, enriching the founders and current operators in currency other than Chinese currency, it's not something that the government would want to see. And, you know, on the margin, there may be some value in having young, young influencers mad at the US government if their platform gets shut down. So taken all together, like dance and I'm certain in conversation, obviously with the with the Chinese government has

decided that they'd rather face down the threat of shutdown than pursue a sale. Hmm. Well, it's so interesting that you note that this algorithm that has been lauded as the secret sauce by ByteDance and by TikTok, that maybe they're overstating exactly what is unique about it. Because we have heard about a sale, the valuation of TikTok has varied from, you know, as little as $30 billion, which is not, does not sound like a little number. And yet, yeah, yeah. Yeah. And as much as 100 billion or even more than that. And a lot of that has to do with the algorithm. One of our our our followers here, one of our community members submitted questions, and I'll say all of you can submit questions as well. We'll keep track of some of those and perhaps ask them during the event. But ask if Mr. Wonderful

of Shark Tank fame might save TikTok who put in an early sort of bid a year ago. But I guess how likely do you think a sale is before January 19th? – I don't think a sale is likely before January 19th with emphasis on the date, because President-elect Trump has also asked the Supreme Court to delay implementation. Again, when Biden signed the legislation last April, there was a high degree of flexibility over the implementation date based on what the president judged and the president staff judged was going on. If they felt like TikTok was making progress and they needed another 90 days, they could

have extended. Not only did they not extend it, they made the date the law goes into effect the day before the inauguration, which ties or would tie then President Trump's hands on the 20th of January. Trump has gone to the Supreme Court and said not strike down this law, but rather delay implementation of this law until I take office.

During Trump's first administration, there was a proposal also to sell TikTok and Oracle was the leading candidate there. Oracle being a tech company that had allied itself with Trump relatively early on. So there is the possibility of a deal if the implementation of the law is delayed until after Trump takes office and Trump wants to use this as leverage for a certain kind of deal making. A big part of the difference between the Biden presidency and the

upcoming Trump presidency seems to be willingness to work not just country by country, which is the norm, but company by company in terms of engagement with foreign companies on US soil. So I do not believe that there will be a sale between now and the 19th. Even the larger figures you listed, the hundred billion dollar figure is relatively, is a relatively small number compared to the Chinese government's desire to keep control of its local internet companies, of which ByteDance is one of the most important. But after the 20th, once Trump takes office, there could be a quid pro quo with China,

where other things are on the table. And at that point, any number of actors could come in and make a play for TikTok's assets. In the old days, sort of the 90s and aughts, Google and Microsoft both thought of themselves as potentially media companies and built various sorts of social media platforms. They could buy their way back in. Amazon, very much in the video content delivery business,

could buy it. But all of that would be in the context of a larger deal between Trump and the Chinese government, not just between a private actor and bike dance. Well, it sounds like you're not putting your eggs in the sale basket before January 19. No, absolutely not. Even at all. And I mean, I guess a big thing that will be the determining factor of this is, of the Supreme Court hearings which are kicking off in just a couple days on Friday. And you outline some of the history of this potential ban on TikTok.

As you mentioned in 2020, the Trump administration first explored the possibility of banning TikTok and then this act that had been signed into law last April was right there sued for that. An appeals government rejected their suit in December. And now we're facing the Supreme Court hearing where we'll have some judgment around whether or not this ban will actually happen. What do you think we can expect

during these court proceedings? Tick-tock has said from the beginning, this law targeting us in particular is unconstitutional. I've always used that language. First Amendment would seem to protect media companies from being forced to sell or shut down. The courts have always been very deferential to the power of the president to make decisions around foreign policy.

And the US makes a very sharp distinction between rules that apply to citizens and to American companies versus rules that apply to international non US citizens and companies headquartered outside the US. It seems to me unlikely, and I will say, I am not a lawyer. I'm just someone who's watching these proceedings out of interest in social media, but it seems to me, given the history, unlikely that the Supreme Court will find

that a company that is as substantially foreign, not just owned, but controlled as TikTok is, can sue the federal government for violation of its constitutional rights. I don't expect the court to say, this law can't go into effect, this law is unconstitutional. The thing that the Biden administration signed last April is null and void. They may kill two birds with one stone by giving Trump the delay he has asked for, but doing it in the context of suspension during further review and so on.

So the one really decisive thing the court could do is to say almost immediately, this law is going into affect January 19th, all app stores have to stop distributing versions of TikTok on Android and the iPhone. But that seems to me to be the least likely outcome. Okay. And it sounds like things are kicking off Friday and that we should have a decision fairly quickly, unless there's some delay on the deadline of January 19. Well, even the delay of a deadline would be a decision of sorts, because the transition from this is a Biden administration preference and this is a law enacted under the Biden administration versus this is a Biden administration preference in law, but it will be enacted under the second Trump administration. That's a very different world for everybody. So even what would seem

to be a fairly milk toast decision to delay implementation of the law in the 24 hours between the 19th and the 20th of January, a massive change in regulatory preferences by the executive branch of the US government is gonna take place, everybody knows it. And a one day delay substantially changes the environment that TikTok is operating. So we have some questions that are coming in from TikTok actually, and one person asked

in response to this conversation around the sale about Oracle and the role they play in TikTok in America specifically. So Oracle has long been interested in TikTok and I don't know how much of the backend Oracle is of TikTok. It is certainly part of their mix. They also make claims about what's on American soil, what's stored in Singapore and so on. So there's, you know, as is often the case with the cloud,

there is a smear of data that isn't amenable to a really crisp definition of X kind of data resides solely in Y kind of system and in Z location. But the operation of the day to day aspects of TikTok and ownership of TikTok are two quite different things. TikTok is a famously dynamic system. And one of the things that, again,

with the algorithm being more tied to essentially observing how users interact with the videos themselves versus the sort of social network piece, TikTok is a better source of cultural novelty than Instagram Reels or Facebook or what have you because you don't get the kind of rich get richer link economy as much on TikTok. So the fact that Oracle's involved in day-to-day operation of TikTok doesn't, I think, address

the questions of both the dynamism of the app itself. The algorithm itself is presumably being tweaked and changed as often happens. But also, TikTok as a source of cultural surprise is at least part of the motivation and worry about TikTok. And that's not accounted for simply by asking who operates the infrastructure. You know, again, this is one of those cases where the stated rationales of all parties tend to give a particular rationale. We are concerned about national security.

We are concerned about freedom of speech. concerned about national security, we are concerned about freedom of speech. Um, but the symbolic position of TikTok as a culturally important part of the American media landscape, um, is in the background of every, every decision being made here. Well, I want to dive into both the national security and freedom of speech elements of it, and maybe we'll start with national security, uh, which of course is the motivation as the U government has has put forward for banning TikTok in the United States. How real is TikTok's threat to US national security? And if you could paint a picture, I guess, of the path we might be on where nothing to actually change about our presence of TikTok in the United States. Yeah, I mean, so TikTok, so if you were concerned about surveillance of American citizens on telecom networks,

which is a perfectly reasonable thing to be concerned about, you would not start with TikTok. We are still discovering the ramifications of an attack by a hacking group called Salt Typhoon, which I think at last count has penetrated nine separate American telecom networks through an old switching technology called SS7. And the SS7 bug is by, I think, Mark Warner's assessment on the subcommittee on basically telecom security, the worst penetration of American telecom networks

by a foreign actor ever in history. And this is going on right now. If you remove not just TikTok from the app store, but you remove TikTok from every phone in America on the 19th, suddenly the entire thing vanishes. You will not have materially changed the security from surveillance in any significant way. I think we've all had the experience or many people have had the experience

of logging into a website being told you have to create a 17 character password that has letters and numbers and symbols and it can't have two characters in a row and etc etc. We do all of that and then a few months later you get a note from that from that organization that says oh our bad our entire database is breached we leaked data on 100 million users right. That is roughly the situation we are in with TikTok versus the SS7 bug, which is to say the wholesale weakness of American telecom infrastructure is so vastly much greater than the ability to track an individual user based on something they might or might not have said on TikTok that if your concern was telecom security, you would not start here. And even if you said, look, the SS7 vulnerability, the American telecom security, you would not start here. And even if you said, look, as a seven vulnerability,

the American telecom networks, we need to keep those safe. That's a completely separate problem. We are also worried about surveillance of app users. If you look at TikTok and you say, okay, well, this is Chinese controlled company, minority of shares, but voting rights, and all the complicated things about where's the data moving.

And then you look at Temu, which is the Chinese shopping app, has about the same number of US users as TikTok. Temu has almost all of the data TikTok has, plus it has your credit card and home address. So if you worried even just limited way which apps were on people's phones, you still wouldn't start with tick tock, you'd start with something like Temma. But again, the fact that tick tock is tied to a bunch of other issues, like a very legitimate concern about both misinformation and mental health effects that come from algorithmic recommendation engines. Again, you would not start you would not start with tick with TikTok. But the symbolism of a Chinese media company operating in an American

environment means that TikTok is getting more attention than either Temu on one side or the telecom networks on the other. And so we have a user who's, excuse me, a follower who's asked, has the government made public any evidence of security threats it has accused TikTok of? And it sounds like there wouldn't be anything that they could specifically point to. Yeah, this is one of the rationales. I think I see something in the comments asking what crime has TikTok been accused of? And of course, this is not a criminal proceeding. The legislation targeting TikTok didn't assert that they'd committed a crime. That would have been an executive function rather than a legislative function.

This, as in said, you know, witness to your question, what, you know, what are the risks and threats that this legislation is nominally seeing on them? The strange thing about national security as a rationale is the courts also have a high degree of deference for setting aside typical standards of evidence. Whether the evidence is merely asserted or whether the evidence is revealed in private but not public, there have been, again, you know, sort of vague observations about people revealing their positions and so on. But if the government has specific concerns about TikTok that are not taken care of by

banning TikTok from military phones, which they've already done, it is certainly not released any data like that. And it's not clear that the courts require the threat to be that specific if national security is invoked and if the entity in question is not run by and is not an American company. In a conversation we had before this, you sort of mentioned the Snowden case and how a lot of ways there were learnings from that and maybe we're not actually leaning into it. Yes, indeed. I mean, this is really the unusual thing about the conversation in the US right now around

information security, worry about being surveilled and so on. When Edward Snowden, who was the NSA contractor who released a bunch of documents about what the NSA was up to back in 2013, when he released those documents, what became very clear was that despite Fourth Amendment protections outlined in the Constitution, the National Security Apparatus, in particular the National Security Agency, was collecting 100% of the signal available to them and collecting and storing and analyzing it. And what's become clear from things like revelations, again, from Snowden about the so-called Five Eyes program, where five different governments, intelligence communities pool and share data.

So if the Australians see someone in America or an American citizen doing something, they will inform American agencies and vice versa for all the participants of this five eyes network. And we know from the SS7 vulnerabilities I mentioned earlier that multiple countries are simply are simply spidering as much data as they can get their hands on. And the plain truth of the Snowden revelations is that we are all under surveillance all the time by multiple governments and presumably by some relatively well-funded non-state actors, and that there's nothing an individual can do about it. You will sometimes see, oh, you need to protect yourself by doing this thing or that thing.

But in fact, the networks we are on are so relatively insecure that no one who isn't themselves an expert in information security can really maintain the kind of security posture that keeps the content of your messages secure, and no one really full stop can keep what's called signal intelligence, which is to say, who's talking to whom when without regard to the content of the messages clear.

So the conversation Post Snowden should have said, I think, if we want to change the way that we think about surveillance, we need to push for regulation and control of governments. But instead, we've stuck with this idea that the individual is responsible for their own information security. The fact of the matter is individuals can't, by doing anything in particular on their phone, protect themselves from wholesale surveillance. And the TikTok worry is very much

about retail level surveillance. I mean, there's an aggregate of data somewhere, but the number of people who use TikTok is not as large as the number of people who use mobile phones in the United States, and that data is also at risk. So the overall surveillance environment is so much more adverse than people are often willing

to deal with or discuss. And I was surprised when the Snowden thing happened, I thought, well, people now understand all competent governments gather and keep all the signal they can get their hands on. And yet a lot of the conversation around this stuff still focuses on what apps are on your phone and what are you doing personally. And it's just not even the right scale at which to be thinking about the problem. And so it sounds like the suggestion by the government that we're trying to protect you by

banning TikTok to protect your security is actually the thinking that is correct. But TikTok is not the actual app that you would tackle in protecting folks. And so I wonder, you know, how much of this is also just connected to the fact that TikTok is a Chinaowned app and how data, the way that TikTok handles data compares to the way other apps and other social media platforms are handling data. Is there any reason to believe that TikTok is working with their data in any different way than say Meta or any other social platform? Well, there's sort of two different ways of thinking about data there.

There's kind of primary use, which is what are you doing with the data you have access to in order to predict or create value for your end users? And then the second is, what downstream uses are being made of the data by people who may want access to it but aren't part of your company. One of the observations made about Chinese companies is that the government of China is very clear that if you have data in China, the government has access to it, period. There's not even the fig leaf of Fourth Amendment protections. But the other thing that's clear is that the fig leaf of Fourth Amendment protections didn't actually stop the NSA from scooping up all of the data, again, in the in the

Snowden revelations. So as a practical matter, when the US government presents a warrant to American telecom companies, by and large, they comply and they comply, and they comply often with requests that are themselves secret. So there's very little way to even measure the amount of data sharing that goes on. We know from another NSA whistleblower that there was a closet in San Francisco, a telecom closet in San Francisco, where whole networks, trunk line traffic went through a second room, which is essentially forking a copy to the NSA. So the difference between American use of data and Chinese use of data in the American context is simply that we don't trust China.

It's not about what controls or constraints are there. It's about a worry that the Chinese and American economic models, cultural models, political models are on a collision course. And the current economic decoupling makes for a really heightened environment of mutual concern. My read of the thing you fed Whitney about, it's a, you know, the concern over surveillance is reasonable, but you wouldn't start with TikTok, is that TikTok is actually useful here to both sides, both to the US and China, because it is highly symbolic act to threaten it,

ban it, require it to sell itself and so on. But it has very little practical effect on the economy, much less on military or national security posture. It's in fact useful to both sides as a kind of prize to fight over without the stakes being very high. And the reason I think that the TikTok ban is not being accompanied by concerns about other kinds of surveillance, concerns about other kinds of algorithmic recommendation, or concerns about other places where the Chinese may be capturing data on American citizens, again, like the shopping apps, is precisely that TikTok provides a way to say, look, we're doing something without actually forcing Congress or the executive branch to act on those other issues.

So I think it's more of. You worked in China for some time and are really familiar with sort of the way the geopolitical tensions play out around tech between China and the United States. How has all of this impacted US-China relationships and how do you think it will continue to impact those relationships going forward if it were to happen? So the big impact here is the success of TikTok in the American market is an example of something that many people in the US thought that they had avoided. China has an incredibly robust and active social media market, including especially WeChat, the ubiquitous app that is closest to the everything app. Elon Musk often says he wants to build in an American context. It has not just social media sharing, but also it's a messaging app like WhatsApp, and it's a shopping app, and you can call a taxi

with it, and etc, etc. But consistently, in the shutdown of Western social media in China, which I was there for part of, I lived there from from 14 to 17, had worked there both earlier and since. When I started teaching at NYU's branch in Shanghai, the freshmen and sophomores I had in my class remembered YouTube being shut down from the country and also Facebook in 2009. So I was dealing with the last class of college students that could remember American social media in a Chinese context.

And over time, they shut down everything. They shut down Instagram, they shut down Pinterest, they shut down Quora during the time that I was there. The shutdown is now complete. There's no American social media of even the narrowest sort, except for a sort of gilded version of LinkedIn that operates there. That import substitution, as it were, allowed the Chinese social media market to explode

because it's the largest internet-connected population in any country in the world. But company after company attempted to spread globally and failed because the amount of customization and and tolerance of censorship that a Chinese company had to build into its product at home often made it a bad product for export and the Chinese were not happy if a company said well we want to offer the same search engine that we have inside China Baidu being the largest. Robin Lee at Baidu said, we'd like Baidu to be a global company, but they were not allowed to offer an uncensored search corpus outside of China because if Chinese citizens

went abroad and searched Baidu and suddenly were able to get material they couldn't get in the country, that would become essentially something that the Chinese government wanted to forestall. So for a long time, it looked like there were two internets. In terms of social media, there were two internets, one was about two thirds of internet size, which is to say, every place in the world that wasn't China and one, the other was the last third, which was the Chinese population itself now closing in on now closing

in on a billion users in a single country. TikTok was the first social media company to have a Chinese parent but to succeed wildly in the West and especially in the United States. And so suddenly the asymmetry of no American companies can operate in China, but Chinese companies can succeed in America had gone from being a kind of theoretical worry that didn't seem like it was going to happen to being a very practical reality. And interestingly, although again, TikTok makes much of their secret sauce algorithm. Their growth in the United States was inorganic, they bought a

company called music.ly musically, that was basically a lip syncing, a lip syncing app that had also been founded in Shanghai. So the merger of two Chinese companies created an app that succeeded wildly in America. And there does seem to be some concern that that asymmetry not be allowed to persist that if the Chinese Chinese are gonna wall off their internet from the rest of the world, which they very much have done, that the rest of the world should have mechanisms for preventing Chinese companies,

particularly Chinese media companies from succeeding abroad. Whether that's good or bad really depends on how much you care about national borders versus international flows. I'm an international flows guy, but I recognize that the people for whom national sovereignty and national borders are a primary concern are alarmed by the possibility of Chinese success in media environments where they don't reciprocate. Well, I think one thing that's even worth noting here is that TikTok doesn't even exist in China, right? There's a they have a sister app that that's there I think called do you enter and maybe I mispronounced

Oh, yeah, yeah, go in and so I think that sort of would lead us to something really interesting connected to their algorithm that I want to get to but before we do that, I'm curious What sort of precedent you think all of this sets for TikTok's operation and other nations? Do you think we might see other places you other nations? Do you think we might see other places? You mentioned Canada. Do you think we might see other countries trying to enact similar bans? So the Canadian strategy is quite interesting, in part because we've actually already seen it affected just a few weeks ago, where they went after the corporate entity but not the software, which is to say they were not proposing, forcing basically forcing bike dance to sell or TikTok and its

parent company to agree to sell a particular flavor of TikTok for Canada. And this is one of those things about America, both as the world's preeminent economy, but also as the source of most globe-spanning internet companies. If Egypt were to attempt to go to TikTok and say, you have to sell to us, there's just not enough money, nor enough value in cutting a deal with the Egyptian government in particular for that to make sense to tick tock. EU regulations are really the only other the EU is the only other

block of users large enough to get that kind of attention. And there's always been a difference between American and EU style regulation. American style regulation, for example, typically treats surveillance concerns as a personal violation, whereas the EU is much more willing to treat it as a kind of social commitment. And because the EU regards a lot of interaction that Americans were manned to individuals users choices, the EU was much more comfortable regulating that on behalf of society, however you however you want to conceive of that. The EU

could implement restrictions on TikTok that also forced it to operate in a kind of subsidiary fashion. But there is not a really a third block of users that is both large enough and commercially important enough to to to bite dance to get their attention. Okay, so the algorithm, it was a thing you've been talking about throughout this conversation. And has the secret sauce that, you know, White Dance suggested is that you say is overblown. But a lot of critics of TikTok have really pointed to this algorithm as a real threat from a place as you noted, thinking about mental health, specifically in young people, and also misinformation. So from the misinformation place,

could you, I guess, talk a little bit about the algorithm and why it might be a threat to the spread of misinformation and how great that threat actually. Yeah, so the interesting thing about the algorithm as it was developed by ByteDance for Douyin in particular is that there is a real concern in China, a concern communicated from the government to Chinese internet companies,

that they minimize the threat of social coordination. Whereas in the American context, there's often concern for information, the spread of information. In the Chinese context, the concern is very much focused on group action. Individual citizens are dissatisfied with individual things all the time in any country of the world. But what the Chinese were especially concerned about is public demonstration, and particularly coordinated public demonstration. Social media, because it dramatically increases the clock speed of coordination, challenges

the Chinese ability to remove things from the internet as quickly as they happen. For people who are watching China during the Omicron phase of COVID, the so-called white paper protests where students went out holding up white pieces of paper saying, you know what should be on here, but we're censored. So we're not, we're not putting, we're essentially we're not putting language on here that would that would directly outrage the government, but everybody knows what we're what we're unhappy about. That stuff spreads so

quickly nationwide that they couldn't intervene fast enough to prevent people from coordinating themselves. As a result, American style virality, which relies on, I said something and my friend said it to their friends who said it to their friends, the sort of six degrees of separation, social amplification.

There are many, many fewer places in China that do that. And the Douyin algorithm in particular, the original bike dance version of the video platform, discovered in that context that they didn't need as much data about what someone's friends were doing if they had more data about how individuals were interacting with the video. And this is in a way a kind of sad outcome.

What it says is your friends matter less when you're doom scrolling and essentially what you are forwarding through, clicking on, watching matters more. But that heightening of an algorithm that watches for more of the kind of addictive consumption of short, short clip video. It's what made Douyin work in the US context. They launched the then, Bike Dance then launched TikTok, the American version. But as I said, very quickly ended up buying Music.ly, order to goose participation. So it was a combination of the algorithm and then having a user base that was already using a similar app that actually got them going. So the algorithm is important. I think what they discovered about concentrating much more on time on video and scrolling behavior and kind of personal behavior for consumption

really did matter in the American context because most people had not yet figured that out. Most companies had not yet figured that out. But it is no longer secret sauce. People have gone to school on the TikTok algorithm. As I said, there was a leak, not of the intimate details of the algorithm, but the basic layout. Anybody who's offering reels or any of the kind of vine descended short form video is doing some version of what TikTok is doing. That having been said, Whitney, to your point about harm, the emotions that get people doom scrolling

are very often fear, anger, rage, disgust. And so there is a real risk that you end up opting into much more extreme content that you started with. TikTok does not seem to me to be the worst actor on that score. And it's certainly not the largest. Instagram has more users than TikTok. Facebook has more users than TikTok in the United States. And of course, the parent company of both Instagram and Facebook, which is Meta, has both of those platforms. So again, if your concern was algorithmic

harm to users, in particular algorithmic harm to teenagers, and TikTok was the kind of opening salvo of this is now how we're going to create a regulatory framework for monitoring what people see. That would be one thing. But A, the TikTok salvo did not presage any other interventions. And B, we're heading into a presidency where the very clear incentive for these companies is to get out of the business of filtering content on behalf of their users almost at all. Mark Zuckerberg of Meta has just said that Facebook is getting out of the business of

fact checking. They're simply going to have community notes and the community notes function is going to be relocated out of California and into Texas. It's a very clear political move. So it's quite reasonable to worry about algorithmic effects of recommendation engines on people's mental health. TikTok isn't the place you'd start. And even if you did start with TikTok, it would only make sense if you were going to have a larger framework for dealing with algorithmic harm, and that framework does not exist. Well, one of our community members who's watching this on TikTok asked,

why haven't US based social media companies been able to replicate the amazing TikTok algorithm? And it sounds like it's available and accessible. And so why are, why are we not seeing that same thing? Yeah. So I go back and forth between wanting to give credit where credit is due on the algorithm. They genuinely did figure out something new, but they figured it out six, eight years ago. A little bit like the formula for Coca-Cola, where the Coca-Cola company makes a big deal about not

letting two people who know the formula ride on the same elevator or whatever. If you added more cinnamon to Dr. Pepper to make it taste more like Coke or something, Dr. Pepper is not flying off the shelf. But the thing that makes Coke Coke now is their reach, their branding, their familiarity, and so on. If another company in 2018 had quickly copied TikTok's algorithm, you might have had competition. But social media very quickly acquires a kind of internal gravity, as it were, that it's very difficult to overcome. Because you're not just copying the TikTok algorithm. You'd be copying the TikTok algorithm, but starting with a very different user base

and starting with a very different set of assumptions. If you look at the top most followed people on TikTok and on Instagram, for example, the most followed person on TikTok is an Italian user, Cabilame, who does these very funny kind of physical comedy videos, sort of the 21st century Buster Keaton. His fame developed almost entirely in that context. He was a very funny guy and he makes funny videos

and TikTok is the ideal place for that. He has, I don. He has 170, something like that 170 million million viewers, every single one of the top 25 on Instagram has more followers than him. So TikTok, although the perception is that it came out of nowhere and has become explosively popular. It is not anything like as popular as Instagram. And monetization, the amount of money that they make from the average TikTok user in a year compared to the money from the average

Instagram user in a year, it's fractured. So TikTok success is surprising because it came into what seemed like a stable social media environment, but in fact it hasn hasn't displaced the major, the other major sources of short form video. There was a long period where any new competitor in the kind of media sharing environment quickly got purchased. It's how Facebook purchased both both Instagram and WhatsApp. Again, under the Biden administration in particular, the idea of rapid consolidation fell out of favor. There was much more antitrust action,

but we could see a world in which there are several more mergers in the social media environment. And at that point, the algorithms that people choose will matter less than the scale they're already operating. Well, I wanna talk a little bit about free speech, which of course is at the crux of this case at the Supreme Court. We'll be hearing in just a couple of days. Bydance has argued that suppressing, by banning TikTok, they're suppressing their First Amendment rights. Is there a real argument here for that? And what door do you think this unlocks for the US government?

Yeah, I mean, there is of course a real argument there. I mean, the First Amendment, you know, it has a set of essentially legal doctrines that have made up First Amendment jurisprudence over the over the centuries. But it's also close to an American religion in terms of people's commitment to what they think the ideals of the First Amendment. And however you interpret the trade off of the First Amendment and constitutional protections versus national security, there is something that should give us pause about allowing the federal government to shut down an American media company. The pernicious threat there is, I think, fairly obvious for anyone who cares about the media. That having been said, the appellate court looked at

argument square in the eye and said, you know what, if you're not an American company, this argument doesn't apply to you. I would be surprised. Again, I'm an amateur court watcher. I am not a lawyer. But I would be surprised if the Supreme Court said, as long as there are enough Americans involved, it kind of doesn't matter what country a company is headquartered in and who the controlling shareholders are,

we will continue to let that media company operate. There are huge implications for companies like Al Jazeera, which famously operate in an international model. Al Jazeera, in a operate in an international model. Al Jazeera, in a way, is nothing if not international. There is no single country that they could slim down to and still continue to be viable, for example. But if the US sets a precedent that the minute you are a foreign owned or foreign controlled firm, none of the reticence about shutting down media environments will apply to you. The downstream effects of that could be fairly significant in

terms of either getting those foreign held media environments to set up American subsidiaries or removing them from American contacts. That having been said, it is also always possible that the result of the hearings on Friday are going to be a much muddier, more sort of one-off deal-making, no precedent is being set here kind of decision that's hashed out in the marketplace rather than in the courts. And as it relates to US owned companies, do you think that a ban would influence regulation on any other social media platform? I don't. You know, it's interesting. Facebook, in a couple of years ago, I think 2022, Facebook hired a lobbying firm to help plant stories that TikTok was a danger to American youth. And Facebook's interest in seeing off the threat of TikTok is, you know, of all the

media companies, TikTok is the worst news for Facebook because anything that makes Facebook's dominance of the, or Metta's dominance of the social media environment seem less than inevitable, as bad for their stock price. But even under the Biden administration, there hasn't been any willingness to say concern about algorithmic recommendation of content should extend to all media platforms and protect all Americans without regard to source. Again, the fact that that that tick tock is a foreign company and has strong ties to a Chinese firm, which has strong ties to the Chinese government is enough that I think people are going to treat this separately. And again, that's a kind of, you know, the United States in a sort of generic political environment model. Under the incoming president, media are assessed in public by how popular Trump is on that

media. He has changed his mind about TikTok, having proposed to ban it four years ago, but now believes that it helped drive youth votes to his campaign and has said, why would I ban a media platform or people like me? So, you know, again, I think the safest bet with the Supreme Court hearings on Friday is to bet against clarity. I don't think that there's going to be a precedent set on Friday that is going to lead to a huge amount of downstream either adjustment or regulation by

American media companies in any direction other than deal making. Well, you know, thinking about just government regulation in tech and we have a community member who asked about data and, you know, essentially saying we choose to be on here. What exactly is the U.S. so worried about in terms of what I think that? Well, again, the stated rationale, national security is not the only rationale here. But I will say if data, and this gets again back to the Snowden revelations, people often believe that data that's individually harmless

is also harmless in aggregate. And that's very often not the case. I'll take Ted as an example, just because the organization that is sponsoring this. Whitney, if I were to give you the timestamp and phone number of every text message between any two people who worked at Ted in the last month, you could tell who was having a romantic relationship that wasn't obvious to the office, right?

A burst of text messages between 10 PM and 2 AM or what have you. That's called signal intelligence. And even if the contents of the message are encrypted or otherwise unreadable, there's a huge amount of value from being able to say, who's talking to whom? Give me a social network map of the organization. If you wanted to be able to reach a particular person in the organization, you could through a signal intelligence map,

identify potential targets who were close to that person and so on. And so the question of what data TikTok has is an, oh, I was watching a Subway server video the other day. I was watching a Get Ready With Me video the other day. That stuff is of relatively minor importance, but things like location, when are you accessing it? Are you accessing it at work or at home?

The other kinds of data you can pull off the phone, depending on what your permissions have been set to, all of that stuff is important. So again, data, you know, data hygiene, controlling the amount that American citizens are surveilled, it's a good idea. And it's not like just because individual uses of the data are harmless. The data in aggregate is also unimportant. But even given all of that, if that were your

actual concern, you wouldn't start with TikTok first. Well, we have just a couple more minutes here and I want to ask a couple questions about creators, at least one question. TikTok has created such a unique audience and a unique community. What happens to this creator economy were the app to be banned in the United States? I mean, this is really a question for Taylor Lorenz. Lorenz wrote a fantastic book called What happens to this creator economy where they have to be banned in the United States? I mean, this is really a question for Taylor Lorenz. Lorenz wrote a fantastic book called Extremely Online that essentially tracks the creator economy

from the early days of the mommy bloggers all the way through up until recently. And the kind of axle of the book is the invention and subsequent destruction of Vine, which was the original highly viral short form video sharing sharing platform. Um, one of the things that comes clear in, in Lorenz's telling is that platforms come and go and people move from one to the other, but that the move is never seamless. And there are always some people who did well in the old environment and do badly in the new environment. And there are some people who had a toehold in the old environment who explode in the new environment. The internet does not lack for places to show

short form video. People who advise the, you know, whatever, four or 5% of content creators who are making real money on on on TikTok have said basically download all your videos, making real money on TikTok have said basically, download all your videos, put them in a backpack, get ready to move to another site. So the creator economy overall, I think will continue to do well. It will be interesting to see if any other app copies TikTok's ability to let people into the mix

with funny videos, even if they're not in some dense social connection with people who have a high amount of traffic. But this is not the only time a social media platform has become a really fertile place for a creator economy and then vanished. We could see another turn of that screw. And again, some people will do well, some people will do will do poorly, I think probably Kabilani will be fine. He makes very funny videos, they are not particularly context dependent. But there are a lot of smaller

creators for whom TikTok has been the absolute making of them, and who may simply not be able to flourish in another environment. But it's, although it's being treated as unprecedented, the sort of change or destruction of previous social media, media platforms and subsequent migration has been going all the way back to the, you know, migration from French to MySpace and MySpace to Facebook. This is, this is an old story. Well, we're just out of time. But before I let you go, Clay, I just would love to hear what you think is the thing all of us should take away from this moment,

regardless of what happens in the next few weeks. You know, I think the most important thing about this is whatever the stated rationales of any of the actors in this space, the symbolic nature of a highly successful media sharing platform operated by a Chinese company or essentially controlled and largely co-invented by a Chinese company is really what's at stake and whatever else the legal decisions may tell you, the

symbolism in the media environment is probably likely to be the largest effect of these decisions rather than legal precedent. Well Clay thank you so much for taking the time to chat with us. Thank you. Great to talk to you, Whitney. Yeah, I learned so much. And thank you all for joining us. Have a good one. That was Clay Shirkey and Whitney Pennington Rogers for our series, Ted Explains the World.

This conversation took place on January 8th, 2025. If you're curious about Ted's curation, find out more at ted.com slash curation guidelines. And that's it for today. Ted Talks Daily is part of the TED Audio Collective. This episode was produced and edited by our team, Martha Estefanos, Oliver Friedman, Brian Green, Autumn Thompson, and Alejandra Salazar.

It was mixed by Christopher Faisy-Bogan. Additional support from Emma Taubner and Daniela Ballarezo. I'm Elise Hue. I'll be back tomorrow with a fresh idea for your feed. Thanks for listening. As a Fizz member, you can look forward to free data, big savings on plans, and having your unused data roll over to the following month, every month. At Fizz, you always get more for your money.

Terms and conditions for our different programs and policies apply. Details at Fizz.ca. Mayday, mayday. We've been compromised. The pilot's a hitman. I knew you were going to be fun. January 24th. Tell me how to fly this thing.

Mark Wahlberg. You know what the difference between me and you is? You're going be fun. January 24th. Tell me how to fly this thing. Mark Wahlberg. You know what the difference between me and yours? You're gonna die. Don't shoot him, we need him. Y'all need a pilot. Flight risk.