How Big Tech Weaponized the Internet and How to Fix It | Tim Wu

23:00Tim Wu:And I think Apple had some kind of product, but it wasn't highly used. And, you know, moving back a little bit, Waze, which originally was called Free Map Israel or something, had this kind of, I don't want to say new age, but very like Web 2.0 or maybe collective vision of how it was going to do mapping. It was all going to be about, you know, user contributions and it's like the people's map. And, you know, it's kind of like to be technical, like Linux to Unix, if that if that lands at all, you know, it's going to be totally different in how it was run. And it looked like it was going to be a real contest between, you know, Waze and Maps or maybe Waze would be bought by somebody to anchor some kind of alternative ecosystem.

23:45Tim Wu:But then one day Google just bought them. and you know so there are two things there number one like okay so now we're back to as opposed to like competitive you know ideas of the future and what's going to happen instead we have like old school john rockefeller techniques so you just buy your competitors that's number one and number two the fecklessness of government in this area was really extraordinary as i say in the book, I once found a staffer and asked what happened. This was an antitrust party of all places. And she said the boss had the following theory, which was that Google is what you use when you want to figure out where you are, and Waze is what you use when you want to figure out where you're going.

24:30Tim Wu:So they aren't actually competitors in her view. And that kind of intellectual bankruptcy captured the era and, to my mind, epitomized our failure to hold the companies to account and force them to compete with each other in a serious way.

24:45Demetri Kofinas:So this reminds me of that era of user-generated content, UGC. Yeah. And it wasn't just in things like Wikipedia or Maps. It was also true in content. When YouTube launched, it launched off the back of user-generated content. And for a period of time, there were many people, myself included, that thought that the future would be user-generated content. But what we found was actually, and this is not a failure of the marketplace, it just reflects consumer choice. It turns out people didn't want user-generated content as much as they wanted highly edited, gate-kept content. They wanted professionally published services, which I think, not to derail us, but maybe we'll have a chance to discuss this, which I think does speak to...

25:32Demetri Kofinas:It's like this interesting line that the platforms look to walk, which is that they're increasingly acting as editors or actually full-blown editors. I mean, they completely control. In fact, they make more editorial choices, I would argue, than the New York Times editorial board or any of the journalists at the New York Times. And yet they claim to be independent, objective, just service providers. They're not putting their thumbs on the scale. So actually, maybe you can speak to that a little bit because that seems to be one of the central challenges in thinking through this stuff. The Twitter file is another great example.

26:09Demetri Kofinas:There was all this outrage about how the government, understandably, that the government was involved. But the larger question was never addressed, which is that a handful of private actors are controlling all the distribution of information. And it's not just about suppression or censorship. It's also about amplification. It's about everything. So is there also, do you think, just fundamentally a lack of understanding by the public about just how captured our speech is?

26:36Tim Wu:Yes. I think you put your finger on this as well. So going back to the original vision, in the 90s, the vision was, well, in the 80s, it was like the Usenet chat. And then in the 90s, it's like the AOL chat room. These are all very dated references. But the basic idea is that the main platforms, I say that because when some of the laws were written in this area, the basic idea is you'd have these open and free discussion forums. And that was the idea of what the internet was. And it was all going to be about people spontaneously talking about what they wanted to talk about and so on and so forth.

27:13Tim Wu:And at some point, and I agree with you, and I think the 2010s is when it really became most serious. The major tech companies went from being, I guess, neutral platforms and became much more like media companies. But they did it in a different way, which is as opposed to like the New York Times does. I mean, the New York Times meets every day, you know, the editors and they decide, okay, what's going to be on the first page? And like, what are we going to run at all? The difference was, and I think you can speak to this too, that, well, first of all, it's like they have a million journalists.

27:57Tim Wu:But second of all, it's like, okay, what are we going to emphasize and de-emphasize? You know, what is even sticking with this platform model, what are we going to make sure people see? How much is that going to be tied to who they are? And so there's this way of taking control of what the message and the flavor of the platform is, yet without without being like a traditional editor. And I think that's what they became. They became the strange hybrid of a platform and a media company. And I agree with you, the takeover of that role really changed this whole idea of user-generated content.

28:36Demetri Kofinas:And further to the point, I'm reminded again, I'm just going to go with these things that just pop into my mind. I remember an interview, a joint interview that Bill Gates and Steve Jobs gave years ago. I don't remember who was interviewing them, but the conversation converged at some point around taste and Jobs' conceded observation that essentially Bill Gates and Windows just didn't have any taste. Microsoft didn't have any taste. And I feel like that's also true in this case, which is that at the end of the day, these platforms have no taste. They view the platforms and the content that's distributed through that as basically the conduit through which to stuff people with paid advertising and grow their revenues by matching advertisers with people that might potentially buy their crap.

29:21Demetri Kofinas:And that's basically what the whole thing has become. It's just become a way of selling ads and nothing else. Yeah.

29:27Tim Wu:I mean, I agree with that. I think you can go either in two directions. One, you are like the original idea and you truly are just an open internet. Maybe the closest to that right now is Reddit, I guess. I mean, I'm not saying Reddit doesn't sell ads, but it's kind of, I don't know how many subreddits they have, but a lot.

29:45Demetri Kofinas:But also selling ads as banner ads or whatever pop-ups is totally different than what these platforms are doing.

29:51Tim Wu:Yes. And then other than that, if you're going to curate, just in terms of having good taste, if you're going to curate, then maybe you want a fully curated product that at least has some dignity. But we're kind of stuck in the middle pretending to be a platform. But as you said, basically taking people's worst instincts and throwing the back at them and trying to get them like honed in on their worst fears. And this is the attention merchants, but basically feeding on the lowest human emotions of all kinds as a business model is like not been the greatest thing for either the internet or our civilization.

30:28Demetri Kofinas:So, you know, I'm sure that there are people out there, I've heard of these people. I actually don't know any of them personally that I think these platforms are great. They like the way they are. Those people might be happy addicts like Kurt Cobain or something, just careening towards their demise. But for those of us who feel like things have gotten worse, my question is, does it matter? I mean, does it matter that people's experience? And when I say, does it matter, I don't mean in a cosmic sense. I mean, from a consumer choice point of view, does it matter at the end of the day? Will there be consequences is first beginning just at the market level, like is there a tipping point where people just check out and the model no longer works?

31:08Demetri Kofinas:And then two, is there a point at which the public discontent grows so large that it becomes a whipping boy for politicians and then regulatory changes become inevitable?

31:17Tim Wu:I mean, I think it depends if we're talking about social media or platform power more generally. So maybe I'll... And you're talking about social media right now. You're right.

31:24Demetri Kofinas:I'm talking about social media and I would throw YouTube in there as well, but platform power extends to things like Amazon and stuff like that. So take that question however you like.

31:31Tim Wu:No, I mean, I'll talk about social media first. I mean, I think much of the damage has been done with social media and we should have, you know, it is a good example of where we just kind of let everything fly out on a belief, which many of us held that if you just sort of let chaos rule, something good will, which I sometimes think is the right answer. But frankly, I mean, this is the subject of attention merchants, a business model, as I said, that feeds on the deepest, darkest emotions that people have, feeds on our worst tendencies, tries to generate addiction, tries to feed, it just is not going to be good for civilization.

32:13Tim Wu:And, you know, we're in the United States, just sticking here. I mean, we're so divided, such terrible people say terrible things about each other. It's just hard to be proud of what we've become. I think the damage is done, you know, and I don't think it, hopefully it doesn't matter where you are politically or ideologically. I mean, you can't think this is what you think a good civilization looks like. So I think that damage is already done. In terms of, if I can turn to platform power, if you will let me go there. I mean, I think the subtle, but then increasingly obvious ways in which platforms dominate the economy and take more and more from every other industry and are copied by industries contributes to a different sort of problem.

33:00Tim Wu:So it's less about just having a failed civilized and more about economic inequality, anger about monopolization, a sense of unfairness, economic insecurity, which historically has tended to lead to a lot of instability in extreme versions, leads to revolutions and war. So I think we've marched pretty far down that path. It's not only the tech platforms. I think we have an economy that has centralized wealth in finance and tech, in particular, and made a lot of other people very resentful about that fact. And I think our politics reflects it. And as I say in the book, if you don't manage to start addressing those kinds of problems, the result, we've seen it in so many other countries, Russia, Germany, the result is increasingly unstable politics.

33:53Tim Wu:So that's a little bit going beyond like, I'm annoyed at Twitter and more into like, yeah.

33:59Demetri Kofinas:So actually, I do want to explore those things because not only do you go in depth with case studies like Amazon in the book, but also stuff like the healthcare system where we've seen all of these private equity rollups that have destroyed private practices that we've all experienced. I've written about this recently as well in some newsletters or housing. So we're going to get to that, but let me rage a bit more about social media as a content creator. Yeah, go for it. Because reading your book gave me cause to think about this a bit more deeply, And it sort of combines the forces of artificial intelligence into this conversation or merges them.

34:33Demetri Kofinas:So as a content creator, my time in this business, not making money for a long time, I wasn't making any money when I started my first blog in 2006, but this goes back two decades now, speaking of feeling old. And I remember that when I started, the same economics that held for content creation held for the broader software industry. It was the zero marginal cost economy. There was no additional cost to gaining additional readers. The cost was all up front in writing and publishing the information. And that has essentially been the model. And it's been coupled largely with advertising. On this show, I like to go to a subscriber model, but for the most part, it's been advertising.

35:17Demetri Kofinas:And even with subscription models, the same economics, some of those core economics, certainly on the cost side, hold. I feel increasingly like, because I've been seeing this recently with the adoption of large language models, that there's a tsunami of content heading toward me. And this isn't just true. At first, you would expect it to see this on the kind of low level crap content out there. But I even see this with folks who are very smart and are able to reproduce smart content by training LLMs on having their voice. So that there's stuff actually now that I might want to read that I don't even have time to read.

35:57Demetri Kofinas:And so I'm just curious if you've thought about this at all. In other words, this problem of just, there's just too much information so that even in theory, if all the information you exposed to was interesting, there would be so much of it that you experience a kind of attention paralysis. And I wonder if you think that there are implications for the business logic of the zero marginal cost economy leading to a kind of breakdown in a sense.

36:22Tim Wu:Yes. And by that, just let me be sure I understand what you're saying. Because you're saying the zero marginal cost of content creation, software creation, both of which are information, both of which can be replicated at almost zero cost.

36:35Demetri Kofinas:Right.

36:35Tim Wu:You're saying that non-human actors are better positioned to take advantage of that.

36:39Demetri Kofinas:So I'm saying two, thank you for asking for clarification. I'm saying two things. First of all, I absolutely believe that the non-human angle actually aligns really well with your Amazon case study. So let's take that for a second. Like in the Amazon case study, you explore how independent retailers selling on Amazon marketplace eventually got squeezed to death on fees. At first, they were able to make a viable living. It was a great model, but eventually they either got squeezed to death because the costs kept growing, all these hidden fees, like when you're checking out at a hotel. Or Amazon launched independent brands, their own brands that competed directly with these and put them out of business.

37:13Demetri Kofinas:Dude, this is the model for the vast majority of influencers on these central platforms. AI is going to push all these influencers out of business. Why? Again, because most of the content creators, interestingly enough, evolved basically to align their content with the objective function of the platform. And they just became basically conduits for ads. And so those people are going to be replaced right away. So I think actually that is one interesting observation, which is it'd be the mass levels of unemployment in the influencer economy, which I'm not pulling out any small violins for those people.

37:47Demetri Kofinas:But then the other observation I was trying to make, and thank you for asking for clarification, is that I know smart people. A good example is a friend of mine who recently started sending me these 20 page papers of his thoughts that were produced by AI by him having spent all this time training in on the corpus of tweets and other things that he's published. And he was sending these to me and they looked very interesting and I wanted to read them. And even if we assumed that they were done perfectly and they were all very interesting, I just didn't have time. I was so overwhelmed by the amount of content.

38:17Demetri Kofinas:And so what I was basically getting as that up until now, the economics of content creation have really been about how do you just get more stuff out there because the platform's rewarded you. That's certainly the case in podcasting. The fact that we publish once a week is something that's actually a negative for this show because we get down rank on visibility. But at what point if it becomes so easy to produce content, and even if we could theoretically assume that all that content was great, where does it all break down? Because at the end of the day, individual human beings have to be in a position to absorb it and their minds can only take in so much information.

38:51Demetri Kofinas:So it seems like there's a breaking point here somehow.

38:53Tim Wu:No, I agree. I mean, I think that one of the questions that I think it will animate the next 20 years are like, what are the real scarcities in our time? I think economics and all of human behavior is driven by scarcity. And sorry to get really big picture, but if you're talking 200 years ago, people don't have enough food to eat or whatever. And that's like the competition for material resources is kind of the primary one. And human labor and the ability to manufacture things, another giant scarcity. And when you innovate around, you invent things, you eliminate other scarcities, it all comes down to time.

39:43Tim Wu:This is just like physics. Like it's really hard. The fact is 24 hours is what we got. You know, it doesn't matter who you are, rich or poor, you know, the week, the day does not expand. I guess you can do without some sleep, but there's only so much of that you can do. And so the limiting factor in almost everything with humans as the, you know, assuming we are still the economic units we're talking about is time. And in that sense, the quality of the content being produced, if a human has to absorb it or read it, that is becoming everything. And frankly, I think the platforms understood this well 10 years ago and have tried everything they can to capture the sort of last resource, which is time.

40:30Tim Wu:And like, I mean, how much time do you spend on the major platforms versus other entities? companies, that's, we can tell where the future lies. Now, to their credit, they're like one of the few companies that really managed to grab some time in our era. And that shows you where

40:47Demetri Kofinas:the competition really is. So I love this. What a great transition. These platforms optimize for engagement and also time spent on platform. And I don't know if you directly connect this observation of the book, but you do talk about economies of scale tipping into diseconomies of scale. And that's where monopoly power becomes so important because now there's no more incremental profit to be gained by expanding your product offering or increasing quality. At this point, you're basically using coercion to extract. And again, the name of your book is The Age of Extraction, to extract profits from your users, and in this case, the internet, who you treat like cattle.

41:29Demetri Kofinas:And it seems to me that we may be reaching a tipping point where the juice is no longer worth the squeeze, or rather that the users of these platforms find these experiences so awful that they independently begin to drop off and no longer voluntarily participate in the marketplace because the incentives are so broken. Does that seem like a plausible hypothesis to you?

41:51Tim Wu:I mean, it is plausible if I were like betting on AI as, you know, the successor to the tech platforms. And I think one of the greatest questions of our time is whether AI fortifies the platform model or challenges and maybe destroys it. If I were betting on AI, it would be that no one's going to spend time on platforms anymore because of the sheer magnitude of junk there. And what they'll want is an AI to find them the stuff they want on the platform. Now, the weird part about that is, you know, the platforms rely on, you were saying, on user-generated content or people being active on them. And so if that's what AIs are harvesting, somehow the model is going to collapse.

42:43Tim Wu:In fact, there's literature called model collapse on just this problem. But I think it is a clear and present danger in the book. And as you've sort of described, I see their business model over the last seven years, maybe 10 years has having evolved away from trying to sort of offer anything like a product and more a model, not unlike a casino, where the whole idea is you just keep people there. You know, you know, time is the most precious resource. And you know, the more people are on your platform, you'll get something off them, whether it's assets, whether it's data, whether it's advertising revenue, whether it's fee revenue, you just want to keep people there and who cares what they're doing.

43:27Tim Wu:And, you know, for a long time, like reading other people's ideas or looking at other people's pictures, I mean, that's been a pretty good model buying things on a giant marketplace. But I think they think the platforms think it's vulnerable. I think the heavy investments in things like sports or movie studios shows that, you know, in a sense, it's almost like a casino where they want to have restaurants where people go and hang out for a while or movie theaters. Like there's a sense you got to do more to keep people on the platforms. But this is a slightly meandering. But I think that is the game right now.

44:04Tim Wu:Now, who has the most loyalty of the most people? And that's obviously a number to be significant and is measured in hundreds of millions of people. And that has become the contest for economic supremacy in our time.

44:18Demetri Kofinas:Just a quick observation, then I'm going to move us to the second hour. I'm pretty sure you talk about B.F. Skinner in the book, because so many of these behavioral algorithms treat humans in this basic scheme of stimulus response. And there have been subsequent research studies that I've read about, one of which comes to mind where they put electrodes in the mouse's brain and basically whenever it hit a certain switch, it released dopamine and the things just hit the switch until they die. And it feels like we've set up a model where it's so extractive that it's creating people who are less productive, less able to meaningfully contribute to a human economy.

44:56Demetri Kofinas:And so it cannibalize itself. It tips the very resource that it depends on into ecological collapse. And it feels to me like one way or the other, whether it's the business model breaking down or the internet platform experience itself becoming so unlivable that people voluntarily choose to drop off or reject the program, that's something we'll have to change. And so what I'm very interested in speaking with you about in the second hour, Tim, is what comes next, both from a public policy and marketplace innovation standpoint, what can we expect to see in the coming years? What are some of the very simple but powerful innovations that could fundamentally disrupt and improve this broken system?

45:39Demetri Kofinas:And how can some of these changes apply not just to the technological platforms that we've spent quite a bit of time talking about so far today, but also the platform model in general in places like, for example, healthcare and housing, which you also explore in the book. For anyone new to the program, Hidden Forces is listener supported. We don't accept advertisers or commercial sponsors. The entire show is funded from top to bottom by listeners like you. If you want access to the second hour of today's conversation with Tim Wu, head over to hiddenforces.io slash subscribe and sign up to one of our three content tiers.

46:14Demetri Kofinas:All subscribers gain access to our premium feed, which you can use to listen to the rest of today's conversation on your mobile device using your favorite podcast app, just like you're listening to this episode right now. Tim, stick around. We're going to move the rest of our conversation onto the premium feed. If you want to listen in on the rest of today's conversation, head over to hiddenforces.io slash subscribe and join our premium feed. If you want to join in on the conversation and become a member of the Hidden Forces Genius community, You can also do that through our subscriber page. Today's episode was produced by me and edited by Stylianos Nicolaou.

46:53Demetri Kofinas:For more episodes, you can check out our website at hiddenforces.io. You can follow me on Twitter at Kofinas, and you can email me at info at hiddenforces.io. As always, thanks for listening. We'll see you next time.