Limitless: An AI Podcast - THIS WEEK IN AI: Grok is Back, ChatGPT 5.6, Meta's New Models, China's Threats

Episode Date: July 10, 2026

🌌 LIMITLESS HQ ⬇️EMAIL US:           [e-mail]:    https://limitlessft.substack.com/FOLLOW ON X:   https://x.com/LimitlessFTSPOTIFY:             https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQAPPLE:                 https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890RSS FEED:           https://limitlessft.substack.com/------So many new AI releases this week! Including xAI’s Grok 4.5 and OpenAI’s GPT 5.6, we discuss their coding, reasoning, and benchmark performance. OpenAI also introduced a new full-duplex voice mode, which lets users speak and listen at the same time, and may preview future AI interfaces.We also discuss Chinese AI labs building their own chips to reduce reliance on NVIDIA and imported hardware. And finally, a few updates on Meta’s new image and video models, ByteDance’s SeeDream image model, and Cloudflare’s payments API for micropayments and AI agent access.sis)------TIMESTAMPS0:00 Grok Is Back2:03 Cheap Intelligence Wins3:51 Cursor Deal Pays Off6:21 GPT 5.6 Arrives10:05 Voice Mode Breakthrough15:17 China’s Chip Push20:32 Meta’s New Visuals25:06 ByteDance Joins Image Gen26:36 Cloudflare Opens Payments28:21 Wrap-Up And Farewell------RESOURCESJosh: https://x.com/JoshKaleEjaaz: https://x.com/cryptopunk7213------Not financial or tax advice. See our investment disclosures here:https://www.bankless.com/disclosures⁠Josh works with Anthropic as a contractor. All views expressed are his own and do not represent Anthropic, its leadership, or its affiliates. Nothing in this episode is investment advice.

I've been waiting so long to say this. Elon is officially back. The first major AI model released from SpaceX AI since they went public, since the $60 billion cursor acquisition is officially live. It's called Grok 4.5, and it's pretty damn good. It ranks around three after Fable 5 and GPT 5.5, but it's really good at coding, and it's astonishingly cheap. It's around half the cost of Opus 4.8, and it's their quickest model yet.

But Elon only got to enjoy this for about a day or maybe even a couple of hours because GPT 5.6 was officially released from open air. And if you're listening to this, you probably have access to it right now, a general release. It's their best bundle yet, extremely good, energetic reasoning, but not quite fable worthy. But they are training GPT6, which is going to be released supposedly a month from now. If you're releasing models, you're probably thinking that's amazing for the AI world. But this comes right at a time where our main adversary in China, has been releasing so many models that are open source and everyone gets to use it. They're super cheap, but they might be considering cutting off the West from using their models. Meanwhile,

Meta, who spent around $25 billion trying to build the best model and have seemingly failed, just released two new models. They're Metamuse image and Metamuse video models. We're going to get into all of this and more on today's episode. There's a lot going on. It feels like nothing happens for a week or two and then five different releases happen all on one day. And that's what just happened. We got new JATGPT. We have a new voice interface, which I love that we'll get into. But I think we do have to start with GROC because, I mean, like you said, Elon is back. The SpaceXAI team, which is now officially formalized, they have a logo, they have their official Twitter handle on X. This is their first release in a long time. And the results are pretty

remarkable. They're kind of classifying this as Opus class on a relative basis compared to Claude Opus 4.8. And it seems to have actually accomplished that on just about all of the benchmarks. In fact, it even exceeds it slightly in some benchmark. for a fraction of the cost. And I think this is one of the more interesting things about this model is just how efficient it is for how much intelligence you get out. And I know that the XAI team or the SpaceX XAI team, they've always been really insistent on maximizing the amount of intelligence per watt of compute. And it seems like this is the first sign of that actually working. For the same amount of power and electricity, they're able to generate tokens that have more intelligence per token

than just about anyone else. Like you mentioned, EJAS, the new GROC model is being trained. It is currently cooking in the oven. They have all the GPUs humming. And when that comes out, I really believe that they're going to have a shot at the top. And we talked about this way in the past, maybe like two or three months ago, about how the SpaceX AI opportunity is really just in their ability to build out infrastructure quickly. And as a result, they actually had too much infrastructure for their use cases. They were selling it off to Amazon. They were selling it off to Anthropic because they just had access and it created this huge business model for them. But now they actually have a use case for this. They're starting to take these GPU.

use, they're starting to train the models, and the results are actually great. And it's like, okay, wow, they were, they were acquired for a couple months, but now here they are, right back on the frontier. And you have to imagine that the next model that's being trained on, however many trillions of parameters is going to be right up there, at least very close to the frontier for what I would imagine is a very, very competitive price. I think the most hilarious part about this is, this is probably Elon's most humble model launch ever. He was super upfront saying, you know, this is Opus 4.8 class grade model. It's not as good as fable, but we are trying. a much larger model. Speaking of size, the size of this model, GROC 4.5, is only around 1.4 trillion

parameter models. Now, if you compare that to Fable 5, that's around 10 to 15 trillion parameters in size. So the fact that they were able to achieve number three on some of these coding benchmarks, like the one that we're showing on your screen right now, Deep Sween bench, is pretty impressive. And you can only imagine what a model maybe five times the size of this could probably be quite competitive. And this brings me to the absolute hell-marry move that Elon and SpaceX AI pulled off a few months ago, or actually less than a month ago, which was the $60 billion acquisition of cursor. Now, if you're wondering, yeah, it paid off massively. Just to bring people up to speed here, at the time of their public IPO, it was pretty much a foregone conclusion that SpaceX had fallen

behind in the race. They hadn't released a meaningful model. GROC 4.2 kind of sucked, and OpenAI and Anthropics seem to be just running away with it. With 4.5, they have officially caught up, and it's because they've been using this moat, this data mode that cursor has. For those of you don't know, for those of you haven't used cursor, cursor is this amazing platform that kind of like pulls prompts across many different models and routes it to the right model at the right time to do the right types of tasks. It's also extremely cost-efficient.

Now, that kind of data is really valuable towards training new models. It's not just about how much compute you put into train a model. it is about how you figure out how that model thinks. And Curse's data mode is really useful for that. So that $60 billion acquisition seems to have paid off pretty damn well. Now, if you're wondering whether I should switch to GROC 4.5, if you're using your LLMs to do Google searchers and do basic research, I would say no. But if you are a software engineer, if you're trying to build a new app,

I would say absolutely yes, give it a go, because this model was designed to build. Some of the demos, which will show shortly, has been pretty awesome. It builds apps from scratch, build simulators from scratch, and it's super cheap and fast to use. I think it spits out around 80 tokens per second, and it's around 17 times cheaper than Opus 4.8 per task. It's pretty awesome. Yeah, I was looking at the SWB benchmarks that they offer, and it showed that the average is solved in about 16,000 tokens versus Opus, which is solved in 67,000 tokens. so it's about a 4.2 times gap on efficiency, which is like really impressive. If you're paying for these tokens and you want something more efficient, this is a model

at least to try to put it in your toolkit to test it out to see where it falls. And hey, if you don't like this one, there's some better news for you because Elon says that SpaceX AI will ship a completely new from scratch foundation model every single month through the end of 2026, which is pretty unbelievable. Like, that's one of those where I'll believe it when I see it because like a foundation model from scratch to build every single month seems like a lot. They're going to have to be training a lot of these in parallel. He has around

500,000 GPUs, which he's so crazy. It is the largest cluster. But he has enough GPUs to feasibly train these models. I think it's, it might be true. Yeah. So, I mean, if you think of it, he's got this giant bakery and he's got like a couple different ovens going at once, all cooking various size desserts, those desserts coming

in the form of grok. So I'm really excited that they're back. I'm excited to see how they evolve over the next few months. It seems like now that they've launched the first one, the next ones are going to come in a pretty rapid iteration and we will have our finger on the pulse. But that is not the only model launch. In fact, that is certainly not the best model launch we had this week because GPT 5.6 is here. Finally, it was kind of here. It was here last week or two weeks ago whenever they released it, but it unfortunately got a little blocker from the government. They've cleared

things up with the government. They said, okay, your model is good to go. So now it is good to go. And if you were listening to this podcast, chances are you can go and use it right now. We are a little early to the news. It hasn't quite released publicly for the world at the time of recording. So We don't have any interesting demos. We will probably record a lot of cool demos, showcasing all the capabilities of the model at some point in the near future. But for now, we have the news that it's out, and this is all three.

They have GPT 5.6, Sol, Terra, and Luna, all broadly available, and they are very much at the frontier. Their costs are, I think, $5 per million input, $30 per million output, so cheaper than Fable, and seems like it's a pretty good model. The results on X seem like people are really enjoying this so far. Yeah, so I was scouring across all the early tested reviews from Theo, from a bunch of other folks. And one thing stood out pretty massively to me, which is Open AI seems to have solved the toughest problem that they've, or the toughest critique that they've been facing, which is taste.

So when you use chat GPT, I don't know if you've used it recently, Josh, it still seems pretty sycophantic. It still sounds like I'm talking to an AI bot versus a human that can actually help me or like a companion that can like help me complete my task. With GPT 5.6, it seems that the feedback is overwhelmingly in favor of taste. It sounds better to use. It's more of a friend to kind of consult with. And that has swayed people to switch from Fable 5 to GPD 5.6. Now, it's important to mention that it isn't really a direct competitor to GPD 5.16, according to these reviews.

Apparently, it still lacks when it comes to certain complex tasks at coding, when it comes a certain complex tasks at reasoning, but it is much, much better. It's a big jump from GPD 5.5. Now, if you're wondering what the fable competitor from OpenAI is, it's called GPT6. And according to Greg Brockman, the president of OpenAI, he said that it's aiming to come out in under a month potentially. So again, we're looking at a new paradigm, a new era right now, where the top model labs are basically releasing a new model every single month. Now, if you compare, I had that to what that looked like, I don't know, a year ago. We had to wait months.

We had to wait months. There were droughts between the news cycles, if you remember, Josh. I don't know how we did it. It was so long between each model launch. It was literally multiple months. It's so funny because I remember we had like a lot of time to test these different models. Remember we attributed a personality to each of these different models? Now they're kind of like converging across like this one uniform kind of face and voice and they're kind of flitting between each other.

I honestly am spoiled for choice. It is the best time alive if you are a builder. If you are someone that's interested in using these models and testing them out, you now get access to all of this stuff for a $20 a month subscription. So I'm super excited to actually get my hands on this and use it. If you're listening to this right now, I probably already am doing this, but you can expect an episode as soon as next week to show you what you can do with it. And this is also great news for people who aren't developers, who aren't builders,

people who just want to use the model. Not only is it more intelligent, but also, You have a new voice model today as well, which is really exciting. So the voice model is something that we did have time to play around with. I was testing it last night. I was chatting with it. And it's pretty good. I love the launch video that they have.

We're showing it on screen. Can we play a clip? Yeah, yeah, please. Play some audio. It's very funny. Let's see. Let's see. Today, we are announcing the all-new chat GPT voice,

powered by GPT Live One, a full duplex conversational partner that is the most powerful voice model ever built. Can you simplify that? Sure can. Can you explain what a full duplex voice model is? Yeah. Imagine a normal call with a friend.

You can listen and talk at the same time. That's full duplex. There you go. And that's the news. It's like you can now listen and talk at the same time and you could talk over this voice in a way that you never were able to before. And I will say, this isn't novel technology. GPT launched this kind of as like a research preview for more developers a month or two

ago where it's kind of live intelligence, but now it's rolled into the actual application. And it's awesome. You can yell at it. You can argue at it. You could have it think about things while you are talking about something else. I was doing a little demo where I was asking it to do some research about something. And then we continued the conversation in parallel while it was researching in the background. And then once it had the correct answer, it came back and it said, hey, by the way, that thing you asked me, here's the answer. And it felt very natural in a way that I've never felt before. And watching this demo and getting the opportunity to try it, I noticed that there were some new visual elements as well. So when I asked like, hey, what's the weather in New York City? I wanted to see

how fast it could pull that up. It was close to instant. And then it shows this like really pretty card on screen that gives you just brief overview of all the weather that's going on. And I'm thinking to myself, wow, this is the first version of whatever their new operating system is going to look like for this hardware device. We have all the rumors in the world about what it's going to look like, what it's going to feel like. And we know that voice is going to be at the core of it. So getting an opportunity to interface with a new voice, I can very clearly see this working well with an AirPods like device. It's like when I'm walking on the street, you could chat with it very naturally. It talks back to you. It has the context of everything around you. And it's really exciting to try.

So I would encourage everyone who has a chat GPT account, give it a shot. It's like a pretty cool technology. This might be a hot take, but I think this is Open AIA's biggest release so far of this here. And it's because I think voice is just the way people are going to communicate with AI going forwards. I can't type as quick as I can think. And in fact, like, I can't remember the last time I typed towards my LLM. I use Whisper AI or I use the microphone and I just kind of like speak at Claude or I speak at GBT. And it comes back with very useful insights. It's a much more natural way of interacting. Now, if you listen to this and you're thinking, why is this significant? Like, I've been using voice the entire time, well, I hate to inform you, but you might have been using a lesser, lower grade model.

Voice typically works on older models, not the cut-to-edge models. It takes a while to kind of like catch up and re-implement with like the GPD 5.5, GPD 5.6. For example, I doubt this is supported by 5.6. So you're already accessing a lower level of intelligence. With this new voice upgrade, you now get access to the higher set of things. It also finally implements a really useful feature that I enjoyed from Thinking Machine Lab. So Mira Murati, who was the former CTO of Open AI, left to set up her own AI lab. She created this model called The Interaction Model under Thinking Machine Labs. And it does a lot of what is shown on this video right now, where you can kind of like interrupt it.

You can kind of like talk to yourself and it knows it's not talking to you. Very natural human type interactions. We've finally got this feature at OpenAI. So now I'm thinking about her lab and what she's built and what her mode is. And I'm thinking, what's the point of using that one? I can just do this on Open AI. And again, it's this trend repeated constantly of these frontier AI labs that are basically looking over the fence, seeing what competitors are doing and saying, you know what? That's pretty cool.

I'll get AI to build this in a couple of weeks and launch it ourselves. Well done from Open AI. Yeah, I like the point of it being the most important release. It very much feels like this is the critical release. Like this is the future way that we are going to interface with AI. and for them to really nail this and to make it a great product. In the mobile app, step one, before it gets moved everywhere else. And you could, I have a feeling that's probably why they used older people.

These are all grandmas. They, three grandmas. It's like, okay, AI is for everyone of all ages. Anyone could speak. Anyone can have a conversation. Now you have an AI to converse with. My fiance's grandma uses this every day. And I'm not joking because I looked at her tracking usage.

Three to four hours on average per day. Whoa. She talks to it. As a friend. as a companion, as a consultant, as a gossip artist when she's gossiping with her friends. It's insane. That's unbelievable. That's high signal too, coming from an older audience because I don't think anyone's really tapped into that just yet.

So this is definitely a place to follow along closely. And my God, we are halfway there to the end of the year when we are going to be getting a preview of whatever this new hardware device is. I am certain that it will be running some form of this voice mode. Next up on the docket, Chinese. AI labs are building their own chips while buying NVIDIAs. EJAS, as our resident Chinese expert, what the hell's going on? Hi, reporting from Beijing over here. Okay, so a narrative that we've consistently talked about on this show is China versus the

US. If you are a new listener, China versus the US is probably the biggest air battle that's happening out there. And the biggest kind of question is, can China catch up to America's frontier of AI models? So far, they've been able to keep up pretty well, but they've lacked in one main thing, which is their chip architecture. They've had to buy and source and rely on Nvidia chips. Until maybe the last couple of months where they announced that companies like Huawei

are building their own AI chips and the government has mandated all Chinese AI labs to use Chinese chips, okay, maybe there needs to be some kind of political resolve there, but now we got announcements for not one, but two AI labs. labs, including the major AI lab from China, Deepseek, that they're developing their own AI chip, according to a bunch of people that are familiar with the matter in order to reduce its reliance on Nvidia and Huawei chips alone. So even just relying less on Chinese manufacturers and focusing on their own AI chips. Now, if I had to guess why an AI lab would do this, it would be twofold. One, you don't want to rely on export controls and restrictions. You know,

Trump or any other future US president might say, I don't want to sell you any more GPU. and your models dead in the water. But number two, the most important one is, I think these AI labs are trying to create a vertically integrated stack. What I mean by that is it's not only good enough to create your own intelligent model right now. You kind of want to own the hardware and infrastructure that it runs on because then you can optimize it specifically for your model. So it could be cheaper, it could work much faster and it could arguably be more intelligent. If you don't believe me, look at what Open AI is doing. Look at what Meta's doing. They're working on their own chips. Open Air is releasing their new Halapeno chip nine months from now. So if they

end up pulling this off, I think we see a world, honestly, a year from now, where a lot of these labs create their own chips and they run it on their own infrastructure, less reliance on Nvidia. I found it really interesting how they've been so quiet about which chips, or if they're even making chips at all, and then within 24 hours, or within even hours of each other, we got the leaks from two places, both Deepseek and Jipu, both building their own custom chips. And I think it's this interesting coordinated effort that we're beginning to see with this rotation, perhaps from open source to more closed source as they feel like they're getting more powerful. So as they get the capability to build these custom ASIC chips, these custom inference chips. And I'll note that these aren't really pre-training chips. These are custom-built accelerator chips similar to that episode that we filmed with Etch about Etch the other day, where these are very purpose-built chips built for optimizing a specific stack.

and it's very clear that China is now in this race and they're trying to get themselves as close to the frontier as possible. And in a way, they're basically now just mirroring the U.S. one-to-one. It's like Open AI has jalapeno. Now Deepseek is going to build their own custom ASIC chip and they're doing it the same place. They're probably using some of the same suppliers. I have to imagine at least the same memory suppliers. So they're now at war in terms of supply chain with the United States. So now traditionally the U.S. versus China has been fought on a software front. This is the first instances in which there's probably some going to be some serious competition on the hardware front, at least as it relates to supply chain and manufacturing these things because the capacity

to do this is so limited. It's a really interesting developments coming out of China and we'll see what they can build with these. Now, we know the process is very long to go from chip architecture to taping out a chip to building a chip and actually producing it at scale, then plugging it in, and then training a model, and then publishing a model. So that takes a very long time. It's going to be while until we get the first models that have actually been trained on these chips. But the seeds are being planted now for a pretty, I would say impressive future from China if they continue on this path. I personally think it's a bigger problem than people are making it out. I remember two years ago when people were like, oh, China would never catch up. And back then it was true. They were like

three years behind. Now that gap is only really six months. And if you can project into the future, let's say that they create a chip that's maybe not invidio level, but it's like 80s, 90% and video level, you could probably train a Fable 5 or Mythos 5 grade model from that. And the Chinese Chippoo founder has literally said, we'll have this ready in a few months, definitely before the end of the year. Now, remember, China isn't the US. There may not be any strict restriction on this. They may use it nefariously. They may use it maliciously. And these models are pretty capable of some dangerous stuff. So I think one of the biggest missteps that we've made, at least on the Western side of things, is imposing these export controls

because it's meant that they have been forced to create their own chip architecture. And, you know, that's what China relied on a lot. I think if we could have just sold them a few more InVdio GPUs, we would have made them more reliant on the architecture. But now it's going to become a heavily siloed thing. And I think, like you said, we're going to lose access to these open source models. I bet you they're going to close source a bunch of the future meters grade models. Yeah, well, I have another company for you that used to be open source and they've gone closed source. So maybe we have a peek into the future because that company comes in the form of meta. Remember them when they were so hell-bent on creating open-source models that were going to change the world, be accessible to everyone.

Well, it turns out that didn't happen. The same way that the metaverse didn't happen, but we still call them meta. Sorry, I feel bad for roasting them, but oh my God, it's so easy to every single time. But anyways, meta is back in the news because they have done something interesting and novel. In fact, well, I will say, I didn't actually realize this was the news. EJ. has filmed me it right before we recorded. I was like, oh, wow, they did release something this week. And that comes in the form of mousse image and mousse spark.

You'll remember the muse models. Oh, I feel so bad roasting them. But like the last experience we have with muse models, that was when everyone's Instagram accounts got hacked. So they're back at it again with a new model. This one now focused on visuals and videos. What is different about this model than the previous one? Because I remember we had that like AI slop TikTok feed thing that kind of came and went

from meta. Is this kind of the next iteration on that model? It's not the next iteration, but it's the, it's the power infrastructure. It's the thing that will power all those future Metaibypes apps and stuff. So the quick TLDR is this is their new image and video model

and it is a lot, lot better than their previous image and video models. Now, the bar, admittedly, was set pretty low back then. I would say this is where GROC Imagine currently is. When I look at this, when I look at GROC Imagine, they're relatively the kind of same grade of quality, which is impressive for Meta because this is a big jump from their previous model.

Now, why I like this model specifically is how it works. There's been this recent trend across any LLM in OpenAI Anthropic with the recent Crock 4.5 model, which is called agentic reasoning or agentic thinking. What that means is they allow the model to, one, access more tools to do its thinking, and two, to think more. They give it more tokens. Meta has one of the largest data centers. out there. So that's what they did with this model. So it's not necessarily an intelligent model.

It thinks more. It uses web search to look at potential images that might make sense for the request that you have. It then gets inspired from those images and it generates its own version. And then it looks at it. And it thinks, hmm, will Josh like this? If it's not good enough to Josh's expectations, I'm going to either edit this using my editing tools or I'm going to completely regenerate this from scratch. So you may have to wait a little bit longer, but you end up getting a much better output. And guess what? It's cheaper because the tokens that they serve from their own data centers are cheaper to serve. So all around, a good engineering effort from meta. I don't think they're anywhere near Fable 5 LLM worthy, but I actually don't think it matters.

I think their main focus should be on niche things like this to do what they're good at, which is social media. Yeah, I mean, it seems like it looks pretty good. This is certainly far from the frontier. And I'm starting to realize now, through all of these releases, how difficult it is to compete at the frontier. It's just like, I mean, I think of Google, I think of meta, I think of SpaceX AI. These are all incredible companies. And they're putting all of their weight behind getting to this frontier. And yet they're still having a difficult time of actually reaching it at the same rate that some other companies like Anthropic and Open AI are doing. That said, I mean, this is a good model for what it is. I love the idea that it has reasoning. I think with Open AI in their most recent

image model that they published, the novel and really exciting thing was that it had reasoning abilities baked in. It didn't just take your text prompt at face value and generate an image. It actually thought about it. It worked through the physics. It had this chain of thought that you can kind of trace to get to a final image and therefore the outputs were significantly better. This is another version of that where it's very smart. So I imagine we get some fun kind of like mobile based optimized features with this where it can recognize details and images really well. It can perhaps like if you have, for example, on this image there is something sitting on the table. You could swap out that element in particular.

I'm sure there's a lot of customization being built around this reasoning ability to understand what it's looking at and kind of reason through the images that you want. And yeah, overall, pretty cool release. Are we going to be using it? Is this something in the daily driver toolkit? Maybe thumbnail creation? No. No, why would you do that when you got Chatsypita English Gen baby?

That's true. Or nano banana. That's true. These are both just significantly better. So there's no real use case for anyone who wants to generate images if you just have a $20 I'm on subscription anywhere else. But it is cool to see more progress coming out of meta. I mean, we have brainwave reading,

and now we have image generating this week from them. So they are pushing the needle forward. We have two lost items on the docket, and then we're going to round this episode out. Since we're on the topic of Image, Chen, I hate to break it to you, but there might be a replacement for GBT image to coming out soon. Seed Dream from BiteDars.

Those are the owners and creators of TikTok, if you didn't know, has come out with their first image model that competes directly, with GPT Image 2. And honestly, if this demo is anything to go by, it is incredibly impressive, but not because of the quality necessarily, but one for the cost. It's super cheap to create and also really easy and intuitive to edit. I mean, look at some of these things. Like I could create websites de novo from scratch using some of these things. I could use different languages,

whatever I want. I could swap in animals for other types of animals. I can go for a certain palette color. It just seems like creator or designer's dream. Definitely one worth experimenting with, one that I'm honestly excited to get my hands on. Yeah, this one is really good. I think anything that by dance releases has been like pretty impressive and this is no shortage. The most interesting thing like you said is the ability to edit things quickly and to kind of workshop things to your own ability. You can separate elements almost as if it's Photoshop. So if you're a designer who's been incredibly frustrated because you can't edit specific features, this is the model for you. You can now basically select any part of the image you want, update it, change it to however you'd like. Really

powerful for customization. So I'd say this probably is less for the average person, more for the pro or the designer who wants to get a little bit more out of the image, who wants a very specific output. And very impressive overall. And the final item, Cloudflare, which is not really a company that you hear often because it's in the back end of the internet, has opened up a new monetization gateway, a new payments API. Now, if you operate any kind of website, it's likely that Cloudflare is the guys behind it that is making sure there's no bots attacking you, that there's no spam accounts, etc, etc. So they can govern visits of a website. The reason why this is important is, let's say you create a cool bit of content. Let's say you create an awesome video on Limitless

that might be for paid subscribers in the future. You can now have an AI agent troll this site and pay a little micropayment. Typically, websites in the internet aren't built for this, but with protocols like X402 from our friends over on the Web3 side of the world, it enables micropayments. You can pay a couple of cents to get access to an article instead of having to pay like a $100 per year subscription and stuff like that. So the reason why I wanted to point this out is I thought it was cool. I think it probably marks the open of a new type of economy, which is agents paying for stuff without you even knowing about it and having the means

to do so. Just thought it was worthy of an update. Yeah, I found it interesting how Cloudflare, they say 52% of crawler requests are now for AI training, which is funny. So half of their crawlers in the world are used just for training better models, which is up from 22% in spring of 2025. And now AI crawlers request content 100 to 10,000 times more than they refer traffic back. So this is kind of built around that thesis, that idea.

It's interesting. It's cool. I think it's a testament to where things are to come. We just had the Cloud Fair publication a few weeks ago that said the internet's now dead. Over half of the traffic is bought. So this is a continuation of that. Okay, well, if half the internet traffic is bought, let's build a payments gateway for them. And that's exactly what this release is.

So just a continuing up the trend on this topic. that. That is all for this week. You are fully caught up in the know with all things AI, Frontier Technology. This is our fourth episode of the week. If you missed any previous ones, go back and check them out. We had one about this company that we love etched. We had the Fable Five return. We had AI finding all the trades that we missed over the last 200 episodes of recording, which was a very fun episode. So there's some good stuff for you to dig into this week if you haven't caught up. If you have caught up, go and enjoy the weekend. Touch grass. It is an amazing weekend, at least in New York, it's looking sunny and beautiful. I hope wherever you are,

the same is true. And yeah, if you enjoyed this episode, don't forget to share it with a friend. Don't forget to rate us on your favorite podcast platform. Any final parting thoughts before we head out? I think the final thing to say is if you love our episodes, if you enjoy our content, and if you would like to support us in any way, we're in the market for sponsors. We're looking for companies and features and products that we're excited to use and talk about to our audience of hundreds, hundreds, hundreds of thousands of people per month. So if that is you or if that is someone that you know that think might be them,

please let them know, please send them a message. Send us a message we're available on X, on email, whatever. We'll leave it in the description below. And yeah, hopefully speak to you soon. Thank you so much for listening and we'll see you on the next one. See you.