Limitless Podcast - Claude Mythos: Anthropic's Leak That's Too Dangerous to Release

Episode Date: March 31, 2026

Everyone's losing their minds over the leak of Anthropic's advanced AI model, Claude Mythos, and its cybersecurity implications. We also touch on competitors like OpenAI’s Spud and Google��s Agent Smith amid the intense race for AI innovation. This leak raises critical security concerns, impacting market reactions and highlighting challenges in the AI industry.------🌌 LIMITLESS HQ ⬇️NEWSLETTER:    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/------POLYMARKET | #1 PREDICTION MARKET 🔮https://bankless.cc/polymarket-podcast------TIMESTAMPS0:00 AI Breakthroughs and Leaks1:20 Claude Mythos Unveiled3:55 Cybersecurity Concerns6:20 Market Reactions7:54 Predictions for AI Models10:03 OpenAI's New Direction13:31 Google’s Agent Smith16:05 The AI Landscape------RESOURCESJosh: https://x.com/JoshKaleEjaaz: https://x.com/cryptopunk7213------Not financial or tax advice. See our investment disclosures here:https://www.bankless.com/disclosures⁠

Three weeks ago, rumors broke that a major AI lab had built a model more powerful, more dangerous and more expensive than any AI model that we had seen before. We didn't know which model lab it would be. We didn't know what the model was called. And then just a few days ago, Anthropic leaked a model called Claude Mythos, which is supposedly more powerful than any model that they've ever built before, a tier above Opus 4.6, which is what we see today. This model is actually so good that it is considered a cyber security threat and can't be rolled out to the public just yet. But it's not just Anthropic that's building a model that is close to AGI like this. Open Air has a model codenamed Spud. Google has a model codenamed Agent Smith.

And there's many more to come this year. But the Anthropic leak wasn't intentional. This was discovered by accident last Thursday, March 26, by a fortune reporter who discovered that Anthropics content management system had a configuration error. And for those who aren't familiar, the content management system, it's how the web server serves files. And within that, there is a config error that leaked nearly 3,000 unpublished assets sitting in this publicly searchable database. Anyone could find them. So two independent security researchers, they went through, they confirmed. And among these files were two blog posts of two models named Claude Mythos and a new tier named Capoebara.

Anthropic immediately removed access to all of this as soon as it came out. out, but then later on, an anthropic spokesperson confirmed that it represents a step change in AI performance and is the most capable model we've ever built. So they confirmed, what we're seeing here is real. Now, the problem is, like this image suggests on screen, we're missing a lot of information. This is a leak that something like this exists, but we don't, we're not sure exactly what. What we do know is that there is the new model tier, EGES like you mentioned, named Capy Barra. It is the new tier that sits above opus. So now the lineup will kind of look like haiku, sonnet,

Opus and then Capoebarra at the top, it doesn't really sound quite right. Maybe that's an experimental name. They might find something better. And then Mythos is the specific model name within that tier. So you can think of Capybara as the weight class and mythos as like the fighter. It's the specific model. Now, according to the leaked documents, this dramatically outperforms cloud opus 4.6 on basically everything, but particularly coding, academic reasoning and the cybersecurity benchmarks. And I think the cybersecurity one is one of the more interesting points here because it's so powerful as cybersecurity that one of the main reasons why they can't release it is to actually prevent people from using it maliciously. Is that right? Yeah, so actually, if we rewind to about a

month and a half ago, Anthropics head of AI security, who is actually a legend in the industry, gave a talk about Claude Opus 4.6 when it had just released, and his talk described how the model was pointed at five to ten very popular open source code bases with no instructions given. And what the model did was very, very interesting. It scanned all those codebases and discovered 500 major security flaws that expert human AI security researchers couldn't discover in decades that they'd been staring and using these exact codebases. So Claude did in a couple of hours what many security researchers couldn't do.

We're talking about millions of compute hours and time spent staring at these codebases testing it. Claude, Opus 4.6, managed to figure this. out. Now, this created a lot of excitement, but also a lot of concern. Now, because these AI security researchers had a good heart, they weren't using this maliciously. But if you did imagine, if this model had been placed to, say, a malicious actor, they could have exploited these for many different reasons. And so these exploits were surfaced and they were fixed. But the question now becomes, what if a more powerful model was made more readily available to anyone or an attacker, for

example, for an adversary that could discover and exploit any future bugs. That's the concern that's around that I have personally around Claude, Mythos, or Capi Barra. This model is supposedly meant to be a tier above anything that we've ever seen before. Apparently, it is amazing at discovering and exploiting exploits. So if it is, let's say, two orders of magnitude, let's be conservative, two orders of magnitude better than Opus 4.6, we could have a real problem on our hands. And so what Anthropic is done now is they've started to slow release. this secret model, Mythos and Capi Barra, two cybersecurity experts first. Why? Because they want them to figure out how they can harden their own defense systems before they publicly release this model

and someone, maybe a nefarious attacker, might use it for unachievable gain. I think it's ironic that the company building, what it describes as an AI with unprecedented cybersecurity capabilities, leaked it because someone misconfigured their blog. And like, the irony there is too strong. And you have to wonder, you have to really ask yourself the question, well, what if the model is so smart that it's leaking itself if it's like poking holes to like let people secretly find it. I don't know. But the one thing for sure is that one, this model is going to be incredibly expensive to run currently at least. That's part of the reason why we're not seeing it now. But the second is it's going to be unbelievably powerful. And the progress that we've had in the last

year is going to probably look like nothing compared to what we're going to get for the next three quarters. The market also very much felt the effects of this because, oh my God, these stock charts look absolutely horrendous. Yeah. Crowdstrong. which is like the major cybersecurity firm was down a couple billion on the news and Palo Alto Networks, which is another similar company that competes in this firm,

also suffered from this. Now, these two charts that I'm looking out right now for these specific companies, Josh, gives me a little PTSD or deja vu because we were talking about this I think four weeks ago when Anthropic released their security review

Claude feature, which wasn't anything to do about Mythos, but basically helped review the vibe code that you produced using Claude. And so cybersecurity stocks dumped again. This is happening seemingly on a monthly basis at this point. Even though these charts are down quite a bit, I'm not sure how concerned the market needs to be immediately

because it appears as if this new model that's coming, this new cybersecurity specialist, is really compute intensive, so much so that it's almost going to be impossible for them to run across all the accounts currently without some serious compression and iteration and figuring out how to run this more. optimally. And it seems like we're starting to see those growing pains, right? It's like as they're

training models like this, as they're running them on their own servers, it's starting to affect the average user. I know sometimes I'll wake up and I'll feel like my opus is running a little bit dumber than it was the day before. And we actually have data that backs this up. Yeah, so basically over the weekend, Claude servers basically went down or were majorly impaired. There were a bunch of different outages. People were reporting very, very reduced quality in their interactions with Claude. And this has been kind of like a repeating trend over the last couple of weeks. And now we might have the answer why. Typically, major AI labs, the last public bit of information that we had was from OpenAI's 2025 run of a major model. They dedicated 30% of their available

compute to a training run. Now, the rumors state that for Claude Mithos, they've dedicated even more. And that's like the major architectural breakthrough that they've made. If they've done that, that might be the reason why we aren't being able to use the best version of code as consumers, because they're too busy using the compute to train the next step or tear in model. I don't know if this is a good or bad thing, but one thing it definitely screams at me is like we need a ton more compute. Big time. And it's amazing to think about how far we've come just in the last three months leading up to this moment here. I mean, when you think about over the winter break is when people really start to take vibe coding seriously. And since then, companies have gone from a very

small percentage of code to almost 100% of code. I mean, this is saying 80% plus of all code deployed is written by cloud code just for Anthropic. It's unbelievable. We started with Opus 4.5, which was released in November. And then Opus 4.6 came in February, which took us from a 200,000 token contacts went into a million. And now, whatever this new thing is, is going to really drive up the coding capabilities in a really big way. And I think it's probably worth checking in on which model is going to be the strongest model, which company has the best model through the end of June, and thanks to Polymarket, we have some interesting stats on this. So the people are betting that Anthropic has a 66% chance of having the best AI model in June, which is huge. And that number

has increased very significantly recently. If you look just back in February, it was Google, who was the heavy favorite, with a almost 80% chance, or 70% chance of having the best model. That has changed recently in a big way, perhaps because of this leak. But, I'm not sure if this is fully up to date and it may be missing some information because we have some news on OpenAI and Google who are planning to release something really important too. And thank you for Polymark for sponsoring that part of the show. But let's talk about Open AI. There's a new code name spud model that's coming.

And this is probably going to be the Mithos competitor. So what does this looking like? Yeah, that's the issue. We don't really know. All of these models, we don't have the specs. We need the specs to talk about him. But there's a few trends or patterns that are happening amongst the hottest, or should I say top two or three AI labs.

We've got Anthropic releasing Metos, which is their AGI or pre-AGI model, a massive, massive leap ahead. Open AI is working on the same thing. They've been secretively working on a larger model. This has gone through a few different names. If you remember, Josh, by the end of the year, I think it was referred to as codename Sprout,

and now it's referred to as Spud. So I don't know if that implies that it's grown massively since then. It's growing. But these models are supposedly meant to be anywhere between 10 to 20 trillion parameter models. Now, for context, the largest models that we currently look at right now is between 1 to 2 trillion. So this is a major order of magnitude larger model.

They're going to be compute-intensive. They're going to be very expensive to serve. So we need to figure out kind of like how to scale AI infrastructure and a bunch of other things. But Open AI's model is codename Spud, and it's meant to be the competitor to Methos. People are anticipating that it might be something like GPT 5.5 or, rather GPD 6. So again, a tier above what we see today. It's going to be advanced in coding,

reasoning, and a lot of the things Anthropics is as well. When I look at this, Josh, personally to me, this seems to be, one, a massive bid to try and leapfrog each other. And number two, maybe try and juice their numbers ahead of a potential IPA. I don't know whether your reaction to this is the same, but that's like my gut reaction when I read news like this. Yeah, it's probably both. They want to juice up things before the IPO, but they also just want to win. And I have some pretty strong speculations just based on vibes of what this is going to look like. I think we've been seeing this recent convergence around OpenAI, particularly on focus and on really dialing in what they're focused on. And we saw a big move last week when they removed SORA.

They totally destroyed Sora. They moved a lot of the teams together. They made their chief of product, the chief of like AGI release. And it appears as if they're building a mega app based on the rumors. So part of the reason why I have a difficult time using Open AI's products is there's kind of spread out everywhere. There's, the Sora app was one, there's Codex, then there's their browser, then there's chat GPT, and there's a lot of different software.

And the same is true with their models, or it was at least, where there was GPT 5.3 codex, and there was 5.3 high, mid, low. There's all these different models that really complicate things and confuse things. With 5.4, they made a singular model. Now, 5.4 does your coding, and it does the reasoning all and one. and what I suspect with this new model code name Spud is going to be the kind of pinnacle of this focus, where I'm hoping they release this with their new application, with a singular model.

So there's one model that is all knowing. There's one application similar to what Anthropic does with the Cloud Desktop app that has all of the functionality under one roof. And I think they're going to probably use this as a point to really lean into that focus instead of distributing this across a lot of different areas. And I'm hopeful that that will meaningfully change opening eye more so than it'll change Anthropic because it actually changes the way that users interface with the product, and it becomes a much better product. Yeah, I think for the majority of last year, I was pretty upset with

the way that Sam and Open AI were focusing on so many different things. I was just like, just focus on creating a really good model. You're being left behind encoding, Anthropics eating your lunch, like, figure this out. And then since their code read of like, what was it, November last year, they've been like reallocating compute, money, data and all their resources to focus on building the best general model and the best coding model. So we're starting to see the fruits of that labor. I have a lot of faith now in Open AI that they're going to produce a really good product that will compete with the likes of Anthropic, which have been eating their lunch. When I look at the last week, it seems like it's pretty negative for Open AI. You mentioned that they killed SORA.

They also killed the $1 billion deal that they had signed with Disney. And they also shut down chat GBT adult mode and a bunch of like consumer shopping apps and their like app marketplace as well. they're just focused on these few things right now. But then the other thing is Sam is also kind of defaulting on a few of the major GPU and Data Center deals, right? So we had the Open Air and Oracle Abilene deal fall through where they couldn't finance it for a variety of different reasons. Then the other thing is they're defaulting on purchasing up to 40% of the world's memory

supply because they haven't still figured out their finances right now. So I think that Open Air is going through kind of like a puberty period where they're figuring their stuff out and where to reallocate resources, but I think they're going to pull through. And it also seems like this is indeed a serious breakthrough. I mean, Sam, in an internal memo that got leaked out to employees, he said things are moving faster than many of us expected. And he called it a very strong model that can really accelerate the economy. That seems like pretty large claims to make internally with employees who are also kind of in the know and aware of what's going on. And I just think that a lot of us who are sitting outside these labs are

not entirely wrapping our head around how much progress is actually about to hit us over the next couple of months with these new model releases. It seems like their step function improvements. And one of the employees from OpenA actually hinted that spud contains a capability that is very different from what we've seen before. So while there aren't specifics, there are clearly a lot of these huge novel breakthroughs incoming, which is worth looking out for. There's one final model release, model leak that we have from Google, who has been doing well kind of chugging along pulling in the background, and this is called Agent Smith. It's a secret AI tool. Do you have any information on this one, EJS? Yeah, so there was like a leaked report from an insider at Google.

Apparently, Google employees are using a new internal tool called Agent Smith that can automate tasks such as coding, according to three people that were familiar with it. The way that this product is supposed to work is within their vibe coding platform called Antigravity, which exists today, but hasn't really had a major upgrade for, let's say, a couple months now, which is like an eternity in the AI world. So they're releasing a new AI model called Agent Smith that is supposed to take a multi-agent approach and use an upgraded version of Gemini 3.1. So it's probably not going to be 3.1. It might be 3.5 or maybe even for, again, another order of magnitude leap up. So what we're seeing here is Google working on an AI coding model competitor to try and catch up to Anthropic and the

likes of Open AIs Codex. You've got Open A.I. trying to reallocate resources and focus on building the best general model and catch up with Anthropic, which they have. at coding. Then you have Anthropic trying to keep these two at bay and make the next order of magnitude up, spending all their compute, but coming at the expense of serving their existing users, which they're adding like a million a day, reporting, you know, closed servers being down and reduced quality of usage. So this is a very, I can like feel the tension in the air between these three companies right now. I don't know what mess is doing. I don't know where GROC is. I'm rooting for them. I hope they catch up. But it seems to be these three major competitors right now that are in

the running for winning this race. They're firing. I mean, in the last 90 days since we started this year to now, we went from 200,000 context windows to a million. We went from these coding assistants to compiler writing agents who are completely capable of writing a very small amount to now over a quarter of Google's production software and 80% plus of anthropic software. Everything we learned this week, the frontier is going to keep moving faster and faster. So we're in for a crazy Q2, Q3, Q4, just a crazy 2026. And as all these things happen, as these IPOs start to happen and they get even more fundraising to deploy these AI data centers at scale. Things are really going to get weird in a hurry, but we will be here to cover it, as always.

If you enjoyed this episode, please don't forget to share it with your friends. Like it on YouTube. Don't forget to subscribe. If you listen on a podcast player like Spotify or RSS, you could rate us five stars there. It's always really appreciated. You guys, any final notes before we sign off for the day? We've been absolutely killing it over on our side. Loads of new subscribers, loads of new listeners.

Thank you guys so much for joining us. And yeah, I have a request because we always like to give out homework at the end of the episode. If you're listening to this and you are an insider anthropic open area or Google and you are willing to give an anonymous tip to our accounts, please spin up an anonymous account on X slash Twitter and DM us. I would love to hear from you. That'd be great. Well, yeah.

Thank you guys for watching. We'll see you in the next one.