Inside the collapse of the internet economy (and what comes next)

This conversation is a practical map of how AI “answer engines” upend the web’s traffic-funded model – and what could replace it. Matthew Prince (co-founder and CEO of Cloudflare) joins Azeem to discuss scarcity, pricing, and why unique, trustworthy content becomes the premium input to AI.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Azeem Azhar?

  Today, I am thrilled to welcome Matthew Prince, the CEO of Cloudflare. It's a company that sits at the very heart of the internet. You might not know of it, but I'm pretty certain that at some point in the last 24 hours, Cloudflare has helped you get something done on the web. The company funnels about a fifth of all internet traffic to us. Now, as we move into the age of AI, Matthew is once again at the heart of thinking about how that technology is going to affect the internet at scale. And that's what we're going to talk about today. Viewers know I love the pipes of the internet, and that means I have a soft spot for this company. I've even got the T-shirt. But that doesn't mean I'm going to give our guest an easy ride. Welcome to the show, Matthew.

• 0:44Speaker 01

  Thanks for having me.

• 0:45Azeem Azhar?

  We're at an inflection point in the way the internet gets used. How broken is the content model that has funded it for the last 25 years?

• 0:58Speaker 01

  Yeah, you know, I think that it's been amazing how well that has worked for the last 25 years. The basic interface of the web for the last 25 years has been search. And Google obviously is the dominant search engine out there. And the way that Google works that you all know is if you type something into Google, you know, especially this or Google of 10 years ago, you got back what was effectively a treasure map, a bunch of links pointing you to search. The search wasn't the act of Google returning the links. You then had to continue the search to find what it was that you were looking for. And in the process of doing that, you generated traffic to a bunch of sites on the web. And Google actually, and some other companies, provided the tools that then allowed those

• 1:41Speaker 01

  content creators, those site owners, turn that traffic into revenue or turn that traffic into ego, just knowing that someone was was reading it. What's happened, and it really started about eight to 10 years ago. It's picked up speed over the course of the last year, is that we're switching from a web where the primary interface is search and search engines to one where the primary interface is AI and what I would call answer engines. You go to ChatGPT and you type a question, it doesn't give you You a list of links to go to, it gives you the answer. And in fact, even Google today, if you go to it and ask a question, increasingly it's putting out the top of the page what they call AI overviews. And those AI overviews are essentially answers to whatever question you have.

• 2:24Speaker 01

  And the problem with that is you're not sending traffic then to the rest of the web. And so if the if the currency that the web was built on, the currency that funded the web was traffic. The problem is that as we switch from search engines to answer engines, that currency is going away. And so we've got to figure out some other way to compensate those content creators to to make the content, which is what fuels fundamentally the web and even what fuels these AI engines.

• 2:52Azeem Azhar?

  As you say, it's something that's been going on for a few years. I think Google was early with this idea of the one box, you know, the one box at the top that constructed an answer and delivered what we wanted, which is we wanted the temperature in X or we wanted to know a simple, the answer to a simple recipe. And this seemed to be, as you say, the evolution. But there is this change because as people move towards ChatGPT, and I think OpenAI claim 700 to 800 million weekly users of ChatGPT. And I think there's some evidence that for every year that you use ChatGPT, you do 8% fewer Google searches. That starts to, as you say, break this relationship.

• 3:35Azeem Azhar?

  But I'm just curious about whether what we're seeing is the the amber warning signal or is it the the red light?

• 3:42Speaker 01

  It's a pretty significant warning signal. The first thing is that the answer isn't to fight progress. Answer engines are a better experience for users. If you think about sci fi films, which predict a lot of our future, when you ask the helpful, you know, robot, what's the temperature outside? It didn't say, well, here are 10 links that you could follow to figure out what the temperature is outside. It told you what the temperature was, right? If you ask it for what's the recipe, it told you what the recipe was. That's the that's the interface that makes the most sense for the future. But what that does is it changes the underlying business model of the web. And so what we're seeing from people who are selling media as they're Primary business is that they're seeing massive drop offs in the amount of ad revenue that they're getting.

• 4:25Speaker 01

  They're seeing massive drop offs in the amount of subscription revenue that they're getting. And as that revenue dries up, it's going to put increasing pressure on content creators. Fewer and fewer people are going to be willing to actually create content if there's not a business model for it. And so what I think is, you know, they're they're sort of the nihilistic outcome. We just say, you know, anyone who's creating content like, you know, starts to death and dies like that. That's a horrible, you know, potential outcome. There's sort of the Black Mirror outcome, which is that, you know, we don't go back to the sort of media of the, you know, nineteen eighties. But we go back to the the media of the fourteen hundreds, which is the Medici's. Where there are five powerful families that unless you kind of are under the patronage of one of them. You don't get to do any research, but instead of families, it'd be five big AI companies.

• 5:08Speaker 01

  Where you've got one that's conservative and one that's liberal and there'll be a Chinese one and there'll be an Indian one. I like those are incredibly regressive outcomes. I think the better outcome is one in which we say fundamentally. Some amount that you're paying for chat GPT, some amount that you're paying for whoever your AI agent is that's out there. As it is giving you value from the content that is on the web. Some amount of that that fee or that or that revenue that the AI companies are generating. That should go back to the content creators and and to me, that just seems like common sense.

• 5:42Azeem Azhar?

  I mean, it's a very appealing notion. It's similar to what we've seen with with, you know, recorded music. You, you know, get the the rights agencies collect revenue from all over the shop and they figure out how to apportion it. I guess one of the things that struck me about what's been happening with with this shift to what you call answer engines. Is that it's unmasked a fundamental leakiness about the internet anyway. You know, the the internet was such that collectively, all of the content amounted to much more than adding, you know, linearly, arithmetically, any single piece of content. And collectively, the value that we got from it was greater than the sum of all the individual bits of value that we as users has got got from it.

• 6:28Azeem Azhar?

  And from all of that. Enormous value, GDP struggled to to count it. We ended up with it being funneled a very small portion of it to to Google. So there was this this issue that there's this enormous wealth of richness out there. And maybe 10% of that value showed up and nearly all of that went to Google. So are we just unmasking a a longer standing problem here?

• 6:53Speaker 01

  If you think about it, though, like, Google has been really one of the real good guys in the web ecosystem. If Google hadn't created all of the monetization tools that we have, I think the web would be significantly smaller than it is today. And so, yes, they have reaped significant rewards, but nobody has actually given back to the open web at the scale that Google has. The irony, though, is that it's sort of like, you know, it's like a Marvel movie. Whoever was, you know, the the hero before becomes the villain, you know, in the next story. And I think right now, Google is at a real inflection point, where they are the ones who are actually saying, we are unwilling to pay for content in this new AI world,

• 7:37Speaker 01

  because we've always gotten it for free. But what's changed, what's changed is that in the past, when Google got the content for free, they gave something back, which was traffic. Now, Google is saying, we want to get the content still for free, but we're unwilling to actually give traffic back. Because the you, and again, it's not of willingness. It's actually the interface doesn't inherently give content back. And so the force that really has to change in order to get this to happen is, Google has to say, the business model of yesterday isn't going to be the business model of the future. Us being the patron of the web is still incredibly important. And we and everyone else who are getting benefit from this for these AI engines that we're creating, we should actually be compensating content creators for that. I'm actually optimistic.

• 8:20Speaker 01

  That one, Google will come around to that perspective, either willingly or they'll be forced to. But two, that when they do, that that will actually be the thing that breaks through the dam. And that we will see across the board, AI companies say, yes, we want to do it. Because almost every single AI company that I talk to says, of course, we should be paying for content. But it needs to be a level playing field. It needs to be something where if we're paying, Google has to pay too. And so, once we get Google to say, yes, we're willing to pay. I actually think that we might actually unlock a huge new business model for the web and for content creators. And potentially unlock really a new golden age of content creation.

• 9:03Azeem Azhar?

  I mean, Google, of course, is one of the players. And I think Google is probably a better player than some of the other AI companies. I think there's some amazing data about the number of times some of these AI models will crawl a website for the number of humans they actually send. And it's sometimes thousands of crawls to send a single person, which might then be monetized. One of the things that strikes me is that the internet is full of a lot of very mid content. I mean, it's got some specialist expert content. It's got some deeply researched material. It's got some very narrow trade information, some very personal information. But a lot of the traditional media industry, I mean, it's in the word media. It's coming from medium, in the middle.

• 9:48Azeem Azhar?

  And an answer engine doesn't mean 27 mediocre sort of similar takes of what you or I or the president might have said. It just needs one. So even if this proposal works, it's going to require some changes to the way media owners and content creators think.

• 10:07Speaker 01

  Absolutely. But I think that what's optimistic to me, if you sort of understand what the incentives are. I think that the incentives of this new model may lead us to an internet that all of us would like better than the internet that we have today. Those of us who have a little bit of gray hair, have been around a while, you know, kind of sometimes long for the internet of old, which was more local, more unique, more creative, more quirky. And that you would stumble across things that were super, super interesting. What gives me hope is it turns out that what AI companies value.

• 10:43Speaker 01

  Is more unique, more local, more original, more quirky. And the best example of this is who is it that has struck the best deal with the AI companies? The answer is Reddit. Reddit and the New York Times, if you add up their total catalogs, have about the same number of tokens in them. The New York Times has obviously been around a lot longer. Reddit, you know, produces more on a daily basis, but it's roughly the same. And yet, the publicly available deal that we know that Reddit did versus the publicly available deals that we know that the New York Times has done. Reddit got about seven times more per token than the New York Times. And the question is why? And I think the answer is exactly what you said, which is, I love the New York Times. Like, I think they're a great media organization. I love the Wall Street Journal. I think they're a great media organization.

• 11:28Speaker 01

  And so is the FT, and so is the Washington Post, and so is, you know, a lot of different newspapers that are out there. But the differences between them are actually relatively de minimis. And you might say, well, you know, the New York Times is more liberal and the Wall Street Journal is more conservative. But it turns out that if I take, you know, OpenAI, ChatGPT, and, and train it on the Wall Street Journal, and then say, okay, give me these stories, but from the perspective of a slightly more liberal publication, they'll basically recreate what is largely the New York Times. And so I think you're right, that in the future, being just sort of samezies with a little bit of a different perspective is not going to be nearly as strong of a business model as being truly unique. But that's what I think is really interesting.

• 12:12Speaker 01

  Like, who has struggled the most in the time of sort of traffic being the currency of media? The answer is something like a small local newspaper in a little town. And, and yet, I think in the future, like, my, my wife and I actually own the local newspaper in Park City, Utah, which is a, which is a resort destination. And we know from the newspaper, we do restaurant reviews on what the best restaurants are. Like, there's nowhere else that you can go in the world that is going to get you that data on what really is the hot new restaurant in Park City, Utah. And yet, if you're someone who likes to ski, or likes to travel, or likes the outdoors, and you're thinking about going to Park City, Utah, if you know that the one-

• 12:57Speaker 01

  Answer Engine out there has the information from the Park Record, our little local newspaper, whereas another one doesn't. You're going to be slightly more likely to subscribe to the one that has that information. And so it's that local information that starts to be the real gold that differentiates one answer engine from another. And if you look going forward, I actually think that the AI companies will look more like a Netflix or a YouTube, where they're competing for original content which really matters than they will like kind of the science projects that we sort of all imagine them to be today. And I actually think, yes, that will change the media landscape, but I think most people, consumers, most journalists,

• 13:40Speaker 01

  will actually say, wow, I wish that I could do more of those local kind of interesting stories and less of the me too, you know, what's happening in the White House today.

• 13:49Azeem Azhar?

  I mean, that's a really interesting dynamic that you have painted. And of course, you're right, your local news and niche news, unless it's industry focused, has really struggled to find a business model since the mid 90s. And I guess, you know, that is the, that's the story behind Reddit, because if you really want a review of how well a washing machine works after five years, however good Wirecutter in the New York Times is going to be, the seven discussions on Reddit will be more real, more visceral, more intimate, people's real experiences. I suppose what that then also does is it changes the nature of many of the intermediaries in, on the web who've evolved to take advantage of SEO. A large part of the web has now become,

• 14:32Azeem Azhar?

  how do we get around Google's algorithms in order to appear on the front page, as opposed to how do we really deliver that insight to a Park City restaurant to the person who's looking for it?

• 14:44Speaker 01

  My even simpler version of that is that, you know, there are publications out there, sort of the BuzzFeeds and Huffington Post of the world, who saw their job not necessarily as what do we write in the contents of the article that we write. It was largely derivative, but they saw their job as how do we write a headline which generates so much rage, or generates so much, you know, fear? That it will actually get somebody to click on it like rage bait headlines that were that were out there. I risk getting on the soapbox. But I think a lot of what's wrong with the world today is that we've just been rage baited into being kind of

• 15:27Speaker 01

  extreme in a lot of different ways. And I think that that is at some level because Google taught us that traffic was the ultimate deity that we should all be chasing after. And again, Google, I think, is a massive force of good on the internet. But Google begets Facebook, which begets TikTok, which feels like we're sort of just spiraling down this attention economy hole that's out there chasing traffic. And yet, traffic has never been a good proxy for value. If there's a car accident on the road, you know, it's not like, you know, just because everyone stopped and looked at it means that that's actually a good thing. We don't want car accidents on the road. What we want are

• 16:09Speaker 01

  actual things that are making humanity better. And what I'm encouraged by is, for the first time in human history, we effectively have a mathematical model for the representation of all of human knowledge. Like you smash Anthropic and ChatGPT, you know, OpenAI and Gemini and everything together, and you you get a pretty good model for what human knowledge is. Not perfect, but again, pretty, pretty darn good. You also then inherently get a pretty good model for where the holes in human knowledge are. And so again, I picture it like a block of Swiss cheese, where there's a lot of cheese that's there, but there's a lot of different holes. Because the business model of the web is going to

• 16:52Speaker 01

  change. No matter what, it's going to change because answer engines are coming and they're better. And as a result, the business model is going to change. What I hope it changes to is one in which we reward content creators that create content that fills in the holes in that cheese, where they're actually doing things where they're making things better. And I my favorite example of this is, like, I flipped to Stockholm, Sweden, in order to meet with Daniel Ek, the founder of Spotify. Almost no one in the history of humanity has compensated creators at scale the way Daniel has. And you might argue that it's unfair in some ways of that.

• 17:30Speaker 01

  But the simple facts are that last year Spotify sent out $10 billion to people who are creating music, which is more than the entire music industry was worth just 20 years ago. And so like you go up there and I was sitting with him and we had this really interesting conversation. And one of the things he told me is, you know, when somebody searches for something on Spotify, if you search for like Taylor Swift, Shake It Off, like they're really certain that they can give you a result. But if you search for something that they may not have anything about, like I want a song to a disco beat about how fun it is to dance with my cat, doesn't exist, right? And they know that it doesn't exist, so they try and give you results, but they also know that those results aren't very good.

• 18:14Speaker 01

  And so they take those searches that have very bad results and they actually publish it back to music creators. And there are music creators that are making tens of millions of dollars a year, creating music for those things that people were searching for. And you might think that that's silly, but when you search for something on Spotify, what you're really searching for is an emotion. You're searching for something that you want some way to connect with. And if the answer is that it doesn't exist and yet some music creator can take that as signal for something that is created and then be compensated for it, I actually think that that's beautiful. I think that's filling in the holes in the cheese. And I think that as we think about what the future business model of the web is, we should be thinking about how can we do more of that? Because again, the world is going to be a much better place if there's less rage bait.

• 18:59Speaker 01

  And if there's more actual content, which is local, unique, and actually filling in human knowledge.

• 19:05Azeem Azhar?

  There's a really interesting idea in there, which starts with the fact that L L M's don't get addicted to dopamine. And so you can, you can put this material in there. And one of my experiences of using the different chatbots is that they are incredibly reasonable. I mean, you can get them to be unreasonable. You can push them in different directions, but they, they are this aggregate of human knowledge and their training and the. Reinforcement learning to get them to behave in particular ways. But it's quite hard to get them to be extreme by the time they they reach us. Now, there's a question as to whether that will always be the case and whether as the companies get more and more powerful, whether the owners won't start to tune them in one direction or or another.

• 19:49Azeem Azhar?

  But there is also this really important point that you you've made, which is we can start to identify more clearly where there are gaps and and and opportunities. And one of the things that struck me when I do deep research queries is that Gemini or ChatGPT or Claude will go out and look at five hundred, six hundred websites. And I would have asked a really simple query that needs needs deep research. And I'm thinking, an academic, a human, wouldn't have gone to six hundred websites to figure out when the bicycle boom in Puria was and what the impacts were. You'd you'd go to two or three good sources. So there's also this this question that these machines are rather greedy and they're a bit a bit dumb, frankly, about how they look at the content that's out there.

• 20:34Azeem Azhar?

  So you described the Swiss cheese model, the sense of filling in the gaps. But one of the things to make markets work, including information markets, is to put in a little bit of scarcity, a little bit of the price mechanism that says you have to pay more for this because it's better. And this is free because it's it's slop. So how do we how do you think about that?

• 20:54Speaker 01

  There are no markets without some level of scarcity. There has to be scarcity in order for a market. We don't have a market for like breathable air, you know, unless you're underwater. Because there's plenty of it and you there needs to there is a market for when you're underwater. Because there's not plenty of it. And that's that's the only time that you have markets is there has to be some some level of scarcity. And so step one, no matter what the business model ends up being, but step one has to be the people who are content creators say enough is enough. We're not going to let you take our content without you compensating it for us. And the crazy stat is that in the next few years, the amount of machine traffic on the web will far exceed the amount of human traffic on the web.

• 21:36Speaker 01

  And there's a real cost to that, like our customers have to pay CloudFlare in order to do that. You've got Pay AWS in order to serve that traffic. If content creators are not only getting no value, but also like having to pay more to service these machines, like just, just brass tax, that's not fair. Someone who is actually benefiting from this should be paying the costs of, of both creating the content, but then also serving that content back out. And that increasingly is going to be these AI companies that are getting the benefit back to it. What I think is important is that the answer should not be, you know, that CloudFlare or any single entity determines what is good and what is bad online. What you want is every AI company to have their own algorithm

• 22:21Speaker 01

  that sort of suggests what it is that is, is going to be valuable that's out there. And the way that I generally think about it is, you know, how unique is the content and how reputable is the content? And it's a sort of two by two on those things. And the more unique and the more reputable it is, based on each different AI company's algorithm, the more valuable it is going to be to them. And so what I, I can picture is you could imagine a world in which the AI companies essentially pay for, we want to be able to ingest this much content per day. And there's a, a price that they pay in order to do that. And then you can imagine almost a discoverability service where it takes the algorithm from OpenAI and other things and says, okay, we've now triaged across all the content

• 23:04Speaker 01

  that was created in the last 24 hours. And here's the, the thing you have to have no matter what. And then they will then basically burn down whatever their daily content budget is in order to get that content back and into, into their systems. That's incredibly simplistic. I think that it's going to be some variation of exactly that that defines how this market takes place. You have a scarce resource, you charge for it, and then you help them discover what of that content that was created is the most valuable. And, and undoubtedly, there will be kind of the SEO of the future that is creating content, which, you know, is exactly what OpenAI wants to think. But if OpenAI has designed their algorithms to say, what we really want is to figure out,

• 23:49Speaker 01

  you know, what the true carrying capacity of, you know, an unladen swallow is, that's going to be the sort of thing that people will then go out and do research on. And, and again, that's filling in the holes in the cheese. And that's advancing in a different way than the article that's like, you know, This one simple trick to eliminate your belly fat, or you can't believe what Donald Trump did today. Like, whatever it is that's just entirely designed to get people pissed off. If instead, we're doing something that actually is answering queries, whether that's actually factual, or creative, or humorous, or whatever it is, like, that is a better world. And that's the one that when the business model changes,

• 24:32Speaker 01

  which it has to, it will for sure change. We should all be thinking, how would we like that business model to look going forward? And then how can we design the incentives to get there?

• 24:43Azeem Azhar?

  There are some models. I mean, financial information is a very good example. You can't get real time data from the CBOE or the New York Stock Exchange without paying a lot. You can't get Platts oil data.

• 24:54Speaker 01

  I mean, Michael Bloomberg made an enormous amount of money just charging for 15 minutes of scarcity.

• 25:00Azeem Azhar?

  Right, he's done pretty well out of all of that. I think the thing that matters for a market like that to work, though, is that the seller and the buyer need to make offers, and they have to come and agree. So that mechanism needs to exist for content creators to say, listen, I'm willing to sell this for a dollar or a cent. And they have to probably do that in real time. Now, we do have some infrastructure to do that. I think real time bidding across ad exchanges is something that's running at trillions of transactions every minute. We've seen these things in the financial markets as well. But it would be a whole new set of skills and capabilities

• 25:43Azeem Azhar?

  for content companies, especially smaller ones. You can imagine the New York Times hiring some economists and being able to do this. But you've talked about the local newspaper in Park City, Utah, and presumably Bean Blossom, Indiana, and everywhere else has got these businesses. What does that layer actually? Look like what you're describing is a fundamental, almost a financial market for this information between the providers and the AI companies.

• 26:12Speaker 01

  I don't know exactly how it will get set up, but I'm pretty sure the answer won't be every content creator figures out exactly what their price is internally and then charges for it. It may not even be that it's sort of a microtransaction. It could be, but it may not be. You know, one version of this is absolutely you have some sort of almost real time bidding through a market maker that is essentially figuring out what the price is. The hard thing is that that works really well for commodities where it's the same in all cases. What I worry about is if we just say a token is a token is a token. That's going to lead to sort of a spammy horrible outcome. Problem number one. Problem number two is we've now just picked who the winners are going to be,

• 26:55Speaker 01

  which are the people who are going to be able to pay the most today. So essentially the, you know, Google, Facebook, Microsoft, Amazon, Apple, that they will be able to say, okay, we'll corner the market by doing this. Ideally, what you really want is a whole bunch of buyers. You want a whole bunch of AI companies that are out there. And you want it to make it so that a new entrant into this has a chance of actually participating. So maybe the price paid is actually a reflection of how many users you have. If you have a lot of users, you pay more. If you have fewer users, you pay less. You also want a bunch of sellers that are involved. You want the local newspapers. You don't want massive conglomerates to exist and you want that to happen between those various things. So it could be that it's sort of a price per transaction.

• 27:39Speaker 01

  The other model that it could be is is much more like what you described, you know, with with the music licensing services with something like a Spotify, where effectively they collect a pool of capital and then they distribute it out in a way which is fair on both sides. And ideally you'll have multiple different players who will say, hey, here's how we distribute it that we think is fair. Here's another company that says, here's how we distribute that you think it's fair. And each of them, you know, tries to negotiate with the big AI companies to get the best possible deal for their, you know, members or for their, for their end customers. But step one has to be, you have to scarcity.

• 28:18Azeem Azhar?

  A quick note, if you want to support us in bringing more of these conversations to the world, please consider subscribing to the show. I know that there's a lot we don't know and we're going to discover this over the coming years. So, but I think it's fun to explore where it might, it might end up. You know, of course, these aren't commodities. You know, when I'm, when I'm buying a barrel of oil or you're buying a barrel of oil, we already know what we're going to get and we know what it might be worth for us downstream. The trouble with an information product is I don't know. And by the time I do know, I can renege, I can repudiate, I can say it wasn't worth what I thought it was worth. So you almost need a mechanism that effectively can rate the likely quality of the content and its potential value.

• 29:02Azeem Azhar?

  And then you have to allow those ratings to be themselves rated. So there's, there's got to be a second layer of trust that gets built on top of that.

• 29:11Speaker 01

  Or you let every buyer define their own rating algorithm. In the bond market, it turns out that yes, you've got Fitch and Moody's, but you actually have the banks that create their own models on what they think the default risks are and, and those sorts of things. At some level, like in this case, like how we're thinking about it at CloudFlare is we're building what is effectively like a pluggable framework where the different AI companies can say, here is my algorithm. And you, you, you think that's going to be like a bunch of work, but they've actually kind of already built it. They think of it as a pruning algorithm because the way these things work is they're building giant trees. But one of the most important things that you have to do in order to make these trees

• 29:54Speaker 01

  not be infinitely large is prune off the information that's already there. A way of thinking of it as back, back to the analogy of Swiss cheese. If you already got cheese somewhere, you don't need more, right? So you just, you, you prune that off. And so you take that pruning algorithm effectively and you give it to CloudFlare and you say, okay, you can see the Content that we can't see, run our pruning algorithm against it, give us a score and then tell us what that score is. And the combination of scoring the content plus scarcity is, I think, the right start of the formula of then what the value is. And then exactly how that transacts from there is, you know, is it a microtransaction per thing? Or is it sort of one kind of bulk fee that you pay and then it gets distributed out?

• 30:38Speaker 01

  Sort of doesn't matter, but what matters is can you score the content? And then is the content scarce? If those two things are the case, then that's going to be a way that you can actually, you know, come up with that market. And the way we think about it is we shouldn't be the ones scoring the content. We should be the ones facilitating the algorithms from all the different AI companies to score that content on their behalf.

• 30:58Azeem Azhar?

  One of the things that would do is it would get around the problem that the Spotify or licensing model runs into. Because one of the problems that model runs into, and I'll just play it back, is that, you know, you look at the pool of revenue that's generated and you figure out how to allocate it to each person who's done a piece of research or written an article and so on. The allocation formula needs to be something that people agree on and it also needs to be auditable. And so OpenAI might say, well, I know I took a thousand of your things, Azeem, but actually they're only worth two quarters. Whereas I took one of Matthews, who's a mate of ours, because he delivers our service and his tweets was worth a million dollars. And that's how we're going to do that. And so that auditing itself is, ends up being problematic.

• 31:42Azeem Azhar?

  Whereas a price discovery mechanism, which, you know, I think what you've suggested, what I've suggested, helps you get there more efficiently, seems like it might work. But then we have this price discovery mechanism and we have this, you know, this efficiency across the market. And then we have what I can only describe as a socialist suggestion from you, which is that smaller companies should sort of pro rata, pay much, much less than bigger companies or perhaps smaller publishers should get, you know, more value than bigger publishers. Because we need, you know, we need them in the soup. So tell me how that works out.

• 32:15Speaker 01

  I think it's actually just, you know, at the end of the day, you have a certain amount of users. They get value from the content. And what you're as an AI company really paying for is on behalf of each of my users, do they get access to this content? It's almost like a subscription fee to the content, which is behind, you know, whatever that's out there. And so, you know, it makes sense to, I think fundamentally, that if you only, if you have an AI that only one person uses, that the value is going to be to accrue to that one person. And so they're going to pay, you know, a relatively de minimis amount. Whereas if you have six billion users, like a Google does,

• 32:57Speaker 01

  then of course they should be paying more because again, they're getting access to the same amount of content, but it's the value is being spread across a much wider population that's out there. And so I guess it's actually very capitalist, not communist at all. I think that the more socialist idea is, you know, is actually on the other side, which is, you know, do you let each publisher negotiate themselves? Or are you effectively creating something which is negotiating, you know, on their, on their behalf? That's where I thought you were going with it. And I think that that is the place that you have to figure it out. Because if you're a small publisher, if you're the Park Record in Park City, Utah, you have no idea how much a restaurant review is worth. And it might be worth, it actually might be worth quite a bit that is out there. And having somebody who can negotiate that on your behalf

• 33:39Speaker 01

  is, is I think going to help the market actually be more fair and more robust.

• 33:43Azeem Azhar?

  If you're New York Times or Reddit, you have the capabilities to, to do that. So Cloudflare sounds like you might act as the agent for smaller publishers in, in some ways, as you, you have in terms of protecting our, our websites for years and years. But is this something that the market can actually, the market participants can figure out? Or is this something where Congress in particular needs to come in with a particular point of view? I mean, a few days ago, President Trump said that AI training is akin to reading a book or an article. And, and many people who are doing research and writing say, well, that's not the case because it's used in a very, very different way. I'm just curious about how you, you think we equilibriate, right?

• 34:28Azeem Azhar?

  We get to this new equilibrium and what help we need from, you know, Congress or the president or anyone else?

• 34:34Speaker 01

  I tend to think that this is going to be something which the vast majority of it is going to be solved through private transactions. And there's not much legislative that needs to to happen in order to in order to allow this to happen. It is technically possible that we think it's actually relatively easy to identify and then block the various AI crawlers that are out there. We go to war every single day. We're a cybersecurity company, first and foremost. We go to war every single day with the Russian, Iranian, North Korean, Chinese hackers. You know, stopping some nerds with a C corp in Palo Alto is not, that's that's child's play.

• 35:17Speaker 01

  You you had the the L L M labyrinth product. Yeah, we can do lots of things to make their lives, you know, very difficult. And again, I don't think we need a congressional action to do it. The one place where I do think that we might end up, and I'm hopeful that it doesn't come to this, where I think that there might be a place for some sort of legislative action is is going back to Google. Because again, Google is in this tough spot where they have sort of said that you have to make a choice. Either you completely exclude all your content from search, or we get to use it for AI, especially the AI overviews that they have. And that gives them effectively an unfair advantage over every other AI company that's that's out there.

• 36:00Speaker 01

  If we are able to get them to voluntarily give that away and split those two things apart. Say, search is different than AI. Then I think that that will actually unlock every other AI company being willing to pay for content. Because now we can actually block Google from using it for AI versus otherwise. So that that's a place that there might be legislative action. Again, I'm hopeful that Google will do the right thing and support the ecosystem before it comes to that. We've also done things at CloudFlaw, because again, one of the things you can do is you don't need legislation. You can just have contract. So we added a license that basically says that site owners have the right

• 36:40Speaker 01

  to say how their content is going to be used by robots and that they can opt into search without opting into AI. And again, that's a contract, which if it's ignored, you know, we can go to court and we can see if it will be will be enforced. And and I think that's right. And again, the best AI rulings that we've had have said, yeah, actually, AI use is a lot like reading a book. With technology, the answer is always yes, it's sort of like that. The problem becomes it's like reading a book, but at massive scale. It's not just about reading one book, it's about reading all the books all at once, right? So it's hard for us to conceive of these things that actually have almost human-like characteristics and qualities, but have machine-like scale and velocity.

• 37:22Speaker 01

  And that's, I think, those are the places where actually it becomes very difficult to figure out how to regulate and to legislate. But I am quite hopeful that it's something that the private markets will be able to figure out, you know, pretty well. In the case of music, like, you know, if you think about the day before Steve Jobs steps on stage and launches iTunes 99 cents a song. The entire music industry is worth about $8 billion. Now, the 99 cents a song didn't turn out to be the right answer. There was a whole bunch of, of, you know, potential legislation that people were threatening to pass. There was a bunch of lawsuits that were there. What ended up actually saving the music industry, though, wasn't any of those things. It was the emergence of Spotify and Apple Music and Tidal and TikTok and YouTube and

• 38:06Speaker 01

  all the things that are now funneling much more revenue into the music industry than has ever happened in the history of the music industry. And so I think that there's a potential for that to happen here as well.

• 38:16Azeem Azhar?

  Let's go back to that point about machine scale. One of the things that so my team and I calculated was that we reckon that at some point this summer, more tokens, which represent about three quarters of a word, were produced by machines talking to humans or to other machines than are produced by the entirety of humanity. And of course, that wasn't the case five years ago. And by the winter, they will be far ahead of us. And, you know, you talked about how internet traffic, which is, you know, has a lot of bots right now, will be vastly dominated by bots. And part of that will be this idea of the agent, thousands of agents that each of us will have across this agentic web going back and forward.

• 38:55Azeem Azhar?

  And it looks like it's a tremendous market expansion because those agents will all require information. Some of it will be private transactional information, other of it will be telemetry, signals, environmental information. So how does this affect the way that we should, we can think about that, the future of that agentic web? I mean, prima facie, it seems like it's going to be a much larger market, but transactions will have to be much smaller.

• 39:23Speaker 01

  I think the answer is going to be different for everything which is, is out there. You know, if you've got A knowledge base for your developer platform. You're going to want that content to be accessible by every agent, which is out there for free. You might even be willing to pay agents to come and look at that, that information because it's marketing data, right? It's a way to help teach people how to use your developer platform. And if you make your developer platform easier to use and, and more programmable, then, then that's good, you know, for, for your business. Today even, you know, we're seeing that while there's been a massive drop in the amount of traffic going to media companies,

• 40:06Speaker 01

  there's still a ton of traffic going from these AI agents and AI systems and answer engines to kind of e-commerce companies. Cause you know, you may ask chat GPT, which camera is the best one for me to buy? But you still have to go buy it from somewhere. And so like, it's going to affect different things in different ways. And our point is really, really simple, which is it should just be up to you, the creator of the service or the content or, or whatever it is. Should be up to you to say, how is your content being used? And some people are going to say, I create things in order to help power the future of agents and, and, you know, the, the future of AI. And, and I'm happy for all of my content to be given away for free.

• 40:48Speaker 01

  Other people are going to say, my business model is to do journalism and I should get compensated for that.

• 40:54Azeem Azhar?

  Quite often, really, really important and valuable information is not going to get bought at large rates by large numbers of people. And you could look at academic research in, in that vein. And what's happened with academic publishing over the last 25 years is that a, a couple of academic publishers have started to absolutely dominate that market. They have incredible profit margins, I mean, enough to make other banks and tech companies feel envious. And so that flow has really, really broken. You know, academics get funded, they have to pay to have their articles, papers included, they have to pay to read them. And that's coming out of money that could go into the public good of scientific research. How do we, how do we get that to work in this model?

• 41:36Speaker 01

  Well, I think it's way easier to get it to work in this model because the reason that that doesn't work is because it doesn't generate much traffic. And if it doesn't generate much traffic, then. In a world where the only thing that determines your value is traffic, then things that are incredibly important research, but don't generate much traffic, are actually kind of, again, you know, pushed off to the side. So I actually think it's the the old web model that has gotten us to the place where because traffic is the only currency, that that's actually driving us away from some of these more valuable things. Those things, if they are filling in the holes in the cheese, like if your if your real goal is, you know, a super intelligence,

• 42:21Speaker 01

  you need to fill in as many holes as possible. And so you're going to be much more willing in that world to pay for truly, you know, unique, trustworthy content. That's on the two by two, that's going to be up in the upper right somewhere. That's going to be incredibly valuable for you. And in the future, that's going to be something which, again, I think that you're going to have much more of a willingness to pay from the AI systems that are there. Especially if you can imagine, like, if it's if it's just five AI companies, then that's not as healthy. But if we have, if we literally have thousands that are out there and you have one that's going to be like, I am going to be the world's best English literature professor as a, you know, as a service.

• 43:05Speaker 01

  Like, of course, they're going to have to read every single academic paper that's out there and they'll be willing to actually pay for that, assuming there's a market for somebody who wants to say, I'm going to hire, you know, English literature as a service, GPT.

• 43:20Azeem Azhar?

  You know, where we are, I think, is a really interesting point in the evolution of knowledge technologies and the way in which we think of the value of of information. And of course, pricing helps to do that, but it still leaves elements of market failure, things that don't get priced, even though they're valuable. So we'll we'll cross that bridge in in time.

• 43:38Speaker 01

  But I think it's it's more likely in a world that is valuing furthering human knowledge as opposed to a world that's valuing traffic. There's a lot of hand waving in what I just said. But we know a couple of things. We know the business model of the web is going to change. We know the things that worked well in the previous business model. We know the things that broke in the previous business model. And I think we can largely, not everyone, but we can largely agree on what we would like a better web in the future to look like. And so, given those things, I think it's up to, you know, anyone who thinks deeply to actually say, okay, it's very rare that you have opportunities where a organization or an institution as large as the internet

• 44:23Speaker 01

  has its business model fundamentally change. And if it does, like, what should we be aiming for it to change to? I can't imagine any question that's more exciting to be thinking about and working on. And I'm, you know, excited about how many of like the top academic game theorists and economists and market designers are helping us try to think through what could a healthier business model of the future of the web look like? And again, maybe it has CloudFlare involved, maybe it doesn't. But I think we should all be trying to think about what it should look like in an ideal case.

• 44:59Azeem Azhar?

  So, last two or three minutes before I let you go to your next appointment, those publishers who sit in the middle, who don't have particularly distinctive content, who don't have a great set of resources to break through, they've got some time to make changes. What are the changes that they should start to make?

• 45:17Speaker 01

  I think go ask your journalists, what's the story that you've always wanted to tell? But your editor says, no one really wants, no one's going to care about that, right? There's so many of those things that are out there. What audience are you serving that is unique to your publication? And how can you go deeper on the things that they think about? I think it'd be amazing if the New York Times did less coverage of Washington DC and did more coverage of what are the best restaurants in Brooklyn? What are the best new kind of off-Broadway shows which are coming up? And so I think that, again, hyper-local, hyper-quirky, hyper-unique content, the stories that nobody else are telling, those are going to be the things that are the most valuable,

• 45:59Speaker 01

  hopefully, in whatever the future business model of the internet looks like.

• 46:03Azeem Azhar?

  It's an absolutely Enormous question. We ended up with this business model almost by accident 25 years ago through some experiments and something stuck and maybe could be a bit more design oriented for the next generation of the internet. Matthew Prince, thank you so much for joining me today. Thank you so much for having me. Thanks for listening all the way to the end. If you want to know when the next conversation is released, just hit subscribe wherever you're listening. That's all for now and I'll catch you next time.