Cloudflare CEO Predicts AI Agents Will Outnumber Humans 1,000-to-1

How are you doing?

Gentlemen, good to see you. I'm doing well.

Great to see you.

Great to see you.

What's new in your world? You made some acquisitions. You made an acquisition. Take us through it.

Yeah, so we bought a company called Void Zero, which makes Vite, which is one of the most popular developer platforms that's there. It's just an incredible team. Evan You, who's the founder, is just a first-class human being, someone who our team is super excited to work with. The team that he's assembled is just great. And I think that this is increasingly becoming the platform that is being used to power a lot of the agents that are running around the internet. And a lot of those agents are running on Cloudflare. And so we think it's just a really natural combination.

How simple is the synergy? Is it you'll funnel those 130 million users who download Vite every month? I think it's 130 million weekly downloads.

Weekly.

Wow. Yeah, that's a lot. Into, you know, preferring Cloudflare. Cloudflare, defaulting to Cloudflare, or is there more synergy under the hood around developer integration, company integration? Like, how are you thinking about this playing out?

Yeah, we continue— we plan to continue to leave it as an open source project and support it and invest in it that way. We want to integrate it closely with Cloudflare's developer platform and make sure that Cloudflare is the best place to run any sort of Vite project that you have. But it'll work in any of the different platforms as as well. And so we just really wanted to make sure that Evan and his team have the support to make sure they could continue to really invest in what was just an incredible platform. And we think that that is going to drive more developers to Cloudflare's Workers platform as well.

What are the headaches for developers these days? I know everyone's concerned about token costs and token budgets. Sometimes that doesn't show up for the developer. It's more like the CEO that's worried about it. But is uptime more difficult to maintain? We've been seeing screenshots of different uptime tracker status pages that have more and more red and yellow on them. Like, what's at the top of the stack and how are you helping?

Yeah, you know, I think that you need a different architecture than we've had to build sort of the last generation of applications for what's coming with agents. If you imagine, you know, there are about 100 million knowledge workers in the United States. If all of those knowledge workers had one agent which was, which was working on their behalf, that would, that would— and each of those was— each of those agents was running in a container, a traditional container like something you'd get at AWS or Google Cloud. The amount of just CPU resources that would be needed to, to run just those, those agents, assuming they just had one agent per person, is about 50% of the total CPU capacity that's generated, you know, by all the different CPU manufacturers that are out there today. And that's just the United States knowledge workers. If you take that to the world, then it's several times, you know, 30 or 40 times the capacity of GPUs and CPUs that are existing today. And so what we really think is that as these agents are creating code, you need a different platform for it. And Cloudflare was built— Cloudflare Workers was built not on containers as much, but on something called isolates, which is a much more efficient technology. And so what we're seeing is that as people are building these agents, as they're using them, it's just a much more natural place to be running them. And it's, I think, why you're seeing more and more of the big AI labs have Cloudflare as the preferred target for where, where their code gets run. You know, OpenAI released a project for their enterprise users a little while ago that again is targeting us. And we want to make sure that with, with things like Vite as, as first-class citizens on Cloudflare, that we can help power that future. Because again, it's not going to be kind of the same system that we, that we built with the hyperscalers. It's going to be something different. And again, I think that we have a really good shot of building that different thing.

So can you help me understand more about the problems of like CPU bottlenecks and then maybe some of the solutions? I'm just thinking about like, I would think bandwidth would be an issue, obviously CPU load, but is there a world where we get to some sort of convention around— maybe it's the robots.txt or just the user agent. And when Googlebot shows up or any other AI system— AI agent shows up, you're just delivering a— it's almost an MCP server, but you're just delivering something that looks more like a JSON package, something that's a little lighter, a little less CPU intensive. Is there a path there to optimization? It feels like we're in the era of, like, squeeze every ounce of performance out of everything. But what's actually going to happen here? Where, like, are we just screwed or is there an opportunity?

No, I think two different problems. So the first is, as you ask these different AI systems to perform tasks on your behalf, what has to happen behind the scenes, especially those tasks get complicated, is that they need to be coordinated in some way. So if you say plan a vacation for me or something like that, what goes on behind the scenes is that there's coordination there. And the best way to make that as efficient as possible, what agents are really good at and these various LLMs are really good at is actually writing code. And so that code needs somewhere to live. And the problem is that if that lives in a container, then you've gotta bring in an operating system, you've gotta bring in all the tooling and everything else. And that's actually extremely heavyweight. And so that's the first place that you've got both a CPU and a GPU bottleneck that's there. And so with, with something like Cloudflare Workers and Isolates, you— it's just a much lighter weight system. And so that means that you can have more agents running on the same CPU infrastructure and be able to provide that. And again, I think that's why a lot of these next generation tools are actually built using our platform. You mentioned something else, which is as these agents go out and interact with the rest of the web, you want to make sure that that is done in the most efficient way. And so if they're pulling down all of the HTML from a web page, and those web pages are designed to be consumed by humans, there's just a lot of cruft on that that isn't necessarily as important. And so some of the things that we're doing are for those customers of ours that want to make sure their content is consumed by agents, want to make sure it's as accessible as possible, possible. We are automatically converting things into Markdown, which is a much simpler, uh, system that, that saves you a ton of tokens. It saves you a ton of processing. It means that your context window, uh, can be— can you can fit more useful information into a, into a context window. So I think there's a ton of optimizations, and we're helping both on the developer side as well as on the content side, making sure that we can, uh, have, have these agents be as powerful as possible and, and get as much done as possible.

John Gruber, total victory. You know he invented Markdown?

I did not.

Grubinator.

Yeah. No way.

Yeah. Isn't that amazing? Yeah.

I mean, it's one of these things that just, you know, it turned out it was ahead of its time, but it's such a key for making sure that we can take information and make it, you know, as compact as possible.

John Gruber created the format for God. It's a funny, funny world we live in.

Talk about— it was last week you guys announced that the threshold had been passed around agent versus human traffic. Talk about that moment. Did it happen sooner or later than you expected? Much, much sooner.

Yeah. I mean, it was— I, at the end of last year, so end of 2025, I said that I thought that we would pass bot traffic. And that's, that's, you know, across the board. So that's like Google's crawler, but also, you know, the new agents which are coming out, I thought would pass human traffic by the end of 2027. About 3 months ago, I revised that based on the traffic that we were seeing to say that it would actually be in the first half of 2027. And so the fact that it actually happened in the first half of 2026 is just, is just, it's just been extraordinary. And it just shows how quickly this, this is, is growing. And the real key here is that I think that if, you know, you or I as humans were researching to go buy a digital camera or something, we might visit 5 websites and do, you know, a little bit of research and some price comparison. You know, you you just watch as you use these agents, they have boundless attention to be able to just go to maybe 5,000 websites to find exactly what you're looking for, the best price, the best delivery, the best service, and everything that's, that's there. And so that's just driving an enormous amount of consumption of, of, of the internet. At the same time, the other thing that's happening is that for a long time, since about 2015, The web has kind of plateaued. There were not— there were, there were more websites that were being shut down than were being created during that time. In the last 18 months, though, we're back to the web growing at a rate which is exponential. And what— and it looks sort of similar to what was happening back in the early 2000s in terms of, in terms of growth of the web. And so I think you're seeing both more sort of consumption of what's online, but you're also seeing more things online. As it goes forward. And so in both of those directions, you know, that's going to continue to just drive, you know, more and more use of the internet. And I wouldn't be surprised if, you know, going forward, say 5 years, that bot traffic will be 1,000 times human traffic online. And we've got to make sure that we make the internet work for that new future.

Yeah, like every year we're just going to add another 9 to the 99.999% of internet traffic is bots. Do you know what the baseline was like pre-AI? Were we at like 1% bot traffic? I mean, I've said—

No, it was more. It was more than that. It was about 20%.

Okay.

For a long time. Yes. And it's, you know, Google was the largest. And then obviously there's a lot of malicious things that run around online. But that was around what it was for. And it was pretty stable over the history of Cloudflare, at least where we could measure it. So since 2010, You know, it was always kind of around that 20% range. And then, and then it's, and then it's grown. You know, it's, it's in the last few years it started growing steadily and then it really accelerated in the first half of, of 2026.

I want to talk about the inference stack, I guess, like we're seeing two things play out, sort of a bifurcation, like a WWDC. Apple's launching on-device inference that's it's not going to be frontier, but it's going to be usable for sure on your phone, on a computer. They can obviously go to their private cloud as well. And then you have like the new NVL72 models. There's stuff in between on Cerebras chip, fast but expensive. Then there's everything from, oh, it runs on commodity hardware, but it's still pretty big. You need a real desktop for it. And I'm wondering about how you see Cloudflare fitting into that? It would be a logical extension to Cloudflare Workers to have inference on the edge, inference in different places. Like, where do you see yourself offering inference, if at all?

Yeah, well, we— it's actually kind of funny. Back in 2022, we issued a press release that there was a graphics chips company that we were going to partner with in order to put GPUs at the edge of our network. Yeah.

That graphics chip company turned out to be NVIDIA.

Yeah. And what was amazing was at the time it was just crickets. Like, nobody cared. Sure. And so we sort of did a little of it, but we hadn't really rolled it out broadly. And what was funny was, and then you fast forward 2 years later, and all of a sudden everyone cared. And so we basically just reissued the exact same press release. And now that's become— that's become a pretty, pretty important part of our business. And so today you can run inference at the edge of Cloudflare. And because of the fact that we're in over 350 cities worldwide, you know, we're within, within milliseconds of the vast majority of the world's population. We become a very natural place for inference to happen. My, my working assumption has always been that about 50% of the inference that happens will be on device, whether that's your, your phone or your, or your laptop or whatever, but that there needs to be some standard protocol where your phone or your laptop or whatever that local thing is can hand those either long-running tasks or larger tasks off next to the network. And so that would be to us. And then if for some reason you need something that's more than that, that it could handle it back to, you know, some centralized data center with, you know, with more capacity than we may have. And I think that that's what— that's what we're increasingly seeing. I think my assumption was a little bit probably off. I think actually I think less is going to happen on device today because I think more and more of the tasks are going to be these long-running tasks where it's not going to be just, you know, what's, you know, what's the, you know, what's the temperature in New York today? Instead, it's going to be something like help me plan a vacation, go, you know, take into account all of these different things, shop for, you know, different hotels in different places, plan all the plan all the, uh, the, the travel between, between the different locations. Here are the criteria that I have. And that might be something that takes, you know, maybe, you know, it's not, not going to be seconds. It might be, you know, minutes or hours or days in some cases.

I feel like you're gonna have long-running agents just for Park City snow forecasting and reporting, tracking like individual—

every—

do individual, individual runs, you know, maybe a network of, maybe a network of drones that are identifying what's tracked out, satellite photos, playing it Optimizing your routes on the mountain.

A lot of stuff.

That's exactly right. Hopefully it snows this year, unlike last year.

So what are the decisions when you're doing inference on the edge? Because it's sort of a hard pitch to say, I'm going to shave off 300 milliseconds, 600 milliseconds when you're talking about a 20-minute workflow or even a 10-second workflow, you're getting me a 3% boost, is that going to make the difference? But are you optimistic that there will be sort of like an in-between state where there will be inference that happens and it needs to happen within the request loop under a second and then the latency matters?

Yeah, well, I think there are sort of 3 different buckets why people are choosing to run inference tasks on Cloudflare. I think the first bucket is exactly what you said. It's latency. And so especially for those things that have human and computer interaction where, where, you know, any delay feels like it hurts you, that people are trying to make sure that they can get the best, best performance possible.

For voice models, I mean, it feels like that has to— yeah, voice models have got to be local. Yeah, I love that. Yeah.

And so, and so that's— and so, you know, again, there's a case for that. I think the second case is that a lot of times for either privacy or regulatory reasons, people want to keep things as close to where they physically are as possible. And so, you know, I think that a lot especially in Europe and otherwise, of places think that, you know, they made a mistake with the internet kind of 1.0 and 2.0 of having everything go back to Ashburn, Virginia, everything go back to the United States. And so I think that there's, there's a real kind of sovereign desire to keep things local. And that's important. And I think the third thing is we actually can be oftentimes much more cost effective because of the fact that, you know, we are this— we are a network. And because of how we're interconnected with networks around the world, bandwidth for us is effectively free. And so it's very easy for us to get things onto our network. And because of where we deploy systems, we often are in places where we don't have to pay for the space or the power, which allows us to then pass those savings on to customers. So it can be significantly more cost effective to use inference with us than it can be in some other, other places. That's somewhat counterintuitive, but, but, but it's sort of the nature of what we've, what we've built and kind of the secret to what's allowed Cloudflare to, to, to, to deliver as much as we have over time.

There's a ton of —ton of debate over data centers, how they get built, how big they should be, what we need, the pushback, regulation, et cetera. How are you—is any of this affecting your business? Are you thinking about how you position your footprint in the real world? How are you processing the evolving discussion around data center construction?

Yeah, I mean, so first of all, I mean, we're going to need more data centers. I think a lot of the discussion is somewhat silliness. The water consumption. I mean, these are closed-loop systems. I mean, a golf course uses more water than, than probably all the data centers in the United States over the course of a year. So I think that— I think, you know, there's, there's, there's a lot of kind of silly concerns that are there. And but there are, but there are real concerns as well. We've got to make sure that there are efficient ways to get power to these things, that it's not taking power from, from other, other systems that the grid can support. Support that. And so I think those are all, all, um, all good considerations. For Cloudflare ourselves, you know, it— we're, um, we're a little bit different, where in, in the case of, you know, if you're an AWS or a Google or, um, you know, a SpaceX, you're building one big facility or a handful of very big facilities and putting a ton of machines in that, that one facility. Cloudflare is different. We have a ton of machines, but we spread them massively all around the world. And oftentimes We want to go into the places where networks come together. And so those, those can be, you know, some of the oldest data centers in the world. And in any of those facilities, we may not have— we may have only hundreds of machines. But over collectively around the world, we've got, you know, what is effectively much larger than any individual data center, which is, which is out there. So I think we're less, less impacted from, you know, the new builds of data centers. I think we're much more impacted by How do we find our way into the existing data centers? And then how do we make sure that the equipment that we're deploying is as power efficient as possible? Because we're often given sort of, here's the envelope that you have to fit your, fit your, your equipment in. And we're often guests of, you know, whoever the local ISP is or, or, or whoever, you know, partner is that we're working with in the, you know, the thousands of buildings that we're in all around the world.

What's new with Italy?

Oh yeah, um, yeah, Italy and Spain, like, it's, it's, uh, it's, it's— so you're fighting Italy, you're fighting France, we need world peace here, come on!

Are, uh, limited due to your—

are limited, yeah, no, no, no Ibiza for me, or the Amalfi Coast. No, it's, uh, it's, you know, I think it's, it's interesting, um, it's been actually kind of puzzling, um press. So all of this stems back. So for those who don't know, in both Italy and in Spain, there have been pretty aggressive tactics to come after either me personally or Cloudflare. And it's all been driven largely by the football leagues, the soccer leagues in those places. And what they're concerned about is, is piracy. The thing that's been ironic about it is like we don't like We don't like piracy on our network either. We don't make any money off of it. It takes bandwidth. It steals resources. And so we have a whole team that's working all the time to shut, shut the pirates down. And, and yet, you know, the pirates are clever. They find ways to use our system. And because, you know, huge percentage of the internet sits behind Cloudflare, you know, there's going to be stuff from time to time that we don't, we don't capture. It takes us some time to catch with leagues everywhere else in the world. So the NBA and NFL and MLB and the Premier League in the UK and others. We work really closely with them to— if they identify a pirate stream for us to pull that down, because again, it costs us money. We don't like it. But for whatever reason, La Liga in Spain and the league in Italy have decided that instead of partnering with us to take this down, they would sue us instead. And so, so again, yes, it does cramp some of my summer plans.

I want to talk about IPOs, life as a public company. There's a few entrepreneurs that are trying to go public this year. You had a very successful IPO. The stock is up more than 100,000 basis points since you went public. Fantastic run.

But advice for— we're putting everything in basis points now because it sounds even better.

I mean, the stock's up 1,000%. It's fantastic. But 100,000 basis points sounds 100,000 basis points is a lot. 110,000 basis points. Who's counting? But anyway, what was your process like? We were talking about the level of float, lockups. What were the hard decisions? Were there things that we're talking about now with these big IPOs that weren't even on the top 10 of your to-do list? What is important to a successful IPO and then running a public company? Company?

Yeah, I think there are a handful of things that are interesting lessons for us. So first of all, I love being a public company. I think that it's interesting. In jurisdictions where you don't have no-fault divorce, spousal homicides are much higher. Which is, I think, kind of an analogy to the difference between private market investors and public market investors. It's really hard to fire your VCs. It's hard for your VCs to fire you. And so as a result, like, actually, it's kind of dysfunctional, whereas public market— like, I love our investors because if I do something stupid, they call me up and they say, that was stupid. I sold your stock. And then we— and we have a conversation. And sometimes they say, oh, actually, that makes sense now. And I bought it back. Right. But I think you can have actually a much more honest, much more real conversation. And, and that's— and that That's felt actually just a lot healthier than what you see in the private markets.

Yeah, it's so true. I mean, everybody that's angel invested in any number of meaningful companies is going to have portfolio companies where you've fully given up on the company and you just try to be nice and help if you can. But at some point, you're just totally disassociated with the investment and you're just like, And that's a reality. But yeah, so it's very healthy to be like, yep, I'm disassociated, like moving on, put my energy elsewhere.

And in both directions, right? I mean, it's just a better relationship. And so I loved the process of going public. I think the thing that we did, like it was an opportunity for us to really kind of retell our story. Like you don't get to reinvent yourself very often, but the process of writing the S-1 was incredibly just an opportunity to sit down and really do it. And we really dug in and told the story in a way that to this day we still refer to parts of that document to sort of explain what is Cloudflare and how do we work. I think that the most clever thing that we did, and this was advice that I got from Ryan Smith from Qualtrics, was he said, you know, he's like, what are you doing about friends and family? And I'm like, we're not going to do it because you can take 5% of a IPO and allocate it to friends and family. And I didn't want to get my aunt, like, over— if the stock went down or something, I didn't want to have to explain that over Thanksgiving dinner. He said, no, no, no, you're thinking about it wrong. Think about the people who, if they owed you a favor, that they could make a meaningful difference in the future of Cloudflare, and then offer them the ability to invest in the IPO. Yeah. And I was like, is that— like, is that legal? He's like, check with your lawyers, check with the bankers, but the answer is yes. And I was like, some people are going to have conflicts. He said, it doesn't matter. Like, even just the fact you offered it, even if they can't do it, will mean that they'll always remember that super fondly. And it was incredibly good advice. And we made this crazy list of all of these people who were like, at some point, that might be an important relationship. For us? And like 75% of them said yes. And they all made a ton of money as a result of it. And it was one of those kind of— Now your aunt doesn't talk to you anymore. Well, yeah, my aunt doesn't talk to me.

Say, I missed out on a 100,000 basis point move. How could you?

I am always comforted that we actually— so we priced it $15 a share. And it went up to— and that's actually the other thing that's interesting. Everyone, like Bill Gurley and I have fights about this from time to time about IPOs. You have a lot of control over how much the pop is going to be. And we sat with bankers, and we were like, listen, we want it to go up about 20%. And they were like, OK, if you want to go to 20%, you price it right here. So we priced it right at that point, and it closed the first day at $18, which is exactly up 20%.

OK, but just to steelman a little bit, is that a function of your business, not to be rude or anything, but there are some businesses that are just like meme stocks out of the gate, and like, they're in some weird thing, and there's some froth, and like, they have less control, or is it always controllable?

Well, and part of that was like, did you think the fair value of your business was 18% or 20% higher than where you priced it? And so you wanted, like, you basically wanted the market to sort of reprice, and you didn't care about not selling?

Yeah, I mean, if we had priced it at like 13, it would have gone up more. More. But the problem, again, you want to play for the long term here. And so again, my fight with Bill is he says that you should do some sort of system where you actually price it at whatever the top price is. The problem with that is if you look at all the companies that have done that, so the Airbnbs and others, that they basically— that's great for your private market investors. That's great for the VCs. That's great for any of the— the founders that are taking money off the table. But you got to keep running the company like the IPO isn't the end. It might be the end for the VCs, but it's not the end for the operators of the company. And you want to be able to kind of build that relationship and build that over time. And so I think, you know, we worked incredibly hard to be very sort of choosy on who are kind of who we gave allocations to. And we went through every single name on the allocation list. And, and I think as a result of doing that, you know, if you look at like who are top 10 shareholders are. It's been incredibly stable over, over time. And they're, and they're, they're folks who give us, again, great advice and, and, and have been really just really, really great to work with. So I'm sure that there are times you don't have it, but I was, I was, I was surprised at how the banks have a really good sense of if you price it here, here's what's going to happen. And in our case, it was, it was, you know, to the, to the penny.

Last question. How are you thinking about long-term planning as a CEO when we're on these exponentials? You know, the example of like agent versus human web traffic, like you have, you had like more data than almost anyone on the planet and your prediction was still off or your original prediction was still off by, you know, 20, 18 months, 24 months, something in that range. So how are you thinking about overall business planning when, you know, model capability is increasing and you're just seeing sort of exponentials everywhere?

Yeah, I mean, I think a couple of different things. So one, I think we're trying to make bets on people who can really function in really dynamic environments. And that for us has been somewhat different than I think a lot of other people. A lot of other people have cut back, for example, on kind of early career hiring. We, you know, Kleffer's about 5,000 employees, a little bit less than that now. But we hired 1,111 interns this summer, straight out of college. And they're amazing. Like, they're just killing it because they are so native to these tools. And again, I think we've always thought of our job as training the interns. But this year, like, the interns are helping train a lot of us in how to do these things, you know, really well. So I think staying super dynamic. And then, you know, from our perspective, just making sure that we've got optionality going forward. So, you know, it's, it's, it has gotten tougher to get equipment today. Memory prices are through the roof. You know, things are harder. But we've, we just have a team that's always constantly trying to figure out how to refine that. And then generally, I think we're, we're trying to be, you know, big adopters of AI, not just in developer land, but also across the entire, the entire business. So we built something called Cloudflare OS. Address, which allows, you know, anyone on our finance team, legal team, you know, accounting team, finance, anyone that's out there at the company to be able to have access to tools and really figure out how their job gets done. The clever thing that we did to seed it was we actually set up this email address. And initially, we told the team that it was this magic AI model. And if they just wrote to the email address, address and ask it, I have this part of my job that I need to get done. How do I get that done? It would— sometimes it would ask some more questions, but it would actually give you a response back. What we didn't tell people was actually it was a team of humans behind the scenes that was initially receiving these emails. And then they were using AI systems to flesh things out. But what they were really doing was recording all of the key jobs to be done within the organization. Because you have to— if you're going to have AI systems that can help facilitate this, you actually have to record what are the steps that people are doing as, as, uh, as a part of that. And so as a result of that, we've been able to seed this, this Cloudflare OS now with the ability to do things like, if you're on the sales team and you have to do like an account plan, you know, you can literally just type /accountplan and then describe what it is that you want to do, and it will output that. And that's made, that's made our team so much more productive. Um, and again, I think it's giving us the flexibility to be able to respond to, you know, whatever comes next.

Very cool. Final, final question. Are we gonna, are we gonna see an exponential increase in the number of interns this summer? Yeah. We're gonna see 10,000 interns.

I don't know. I mean, it's, I, I, I, I, we, we've gotta find a place to put 'em all. And so it's, I mean, our, like our office in Austin, like we've, it's a big space in Austin. We've run outta space there cuz we have so many interns there. It's, um, I just, I just think that there, I think a lot of companies are doing this wrong where they're saying like, we're not going to bet on the new people coming out of school because AI is going to replace them. I mean, these kids know how to use AI better than anyone else. So we're going to bet incredibly heavily on that. And so far it's paying off. I mean, I think the person who eventually replaces me might literally be one of the interns today because they're just so, so, so smart.

Smart. It's happened, it's happened before. I mean, I don't know if Satya Nadella was an intern, but he has that famous video of him as a product manager demoing Excel. And yeah, oh yeah, nothing like a company man.

I remember, I mean, I, I was on a bunch of, of calls with Satya, um, before he was, before he was CEO, um, and he was, he was just a product manager, you know, at that. So I think just if you, you can find really great people and, uh, and be able to move on.

You're the GOAT as well. Yeah. You know, whenever you put up a 100,000 basis point move in your stock over just a couple of years, you're in contention for GOAT, I say. Always a pleasure, Matt. Always a fantastic time. Thanks for jumping on.

Good to see you guys.