Should We Be Scared of Anthropic's Mythos?

Anthropic has formally announced their most powerful model ever, one that makes Opus-4-6, just a couple of months old, feel of the past. And yet they're not releasing it to the general public. In fact, the entire discourse they're surrounding it with has some people feeling nervous or even scared. Today we're going to unpack what is actually going on and whether that feeling of fear is the right one or not. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. All right, friends, quick announcements before we dive in. First of all, thank you to today's sponsors, KPMG, Blitzy, Section, and Mercury. To get an ad-free version of the show, go to patreon.com/aidailybrief, or you can subscribe on Apple Podcasts. Remember that it is just $3 a month for those of you who want to cut out the ads. Click the sponsors tab or shoot us a note at sponsors@aidailybrief.ai. And while you're there, you can find out about all the other things going on in the ecosystem. A couple quick ones to mention. Enterprise Claw Cohort 2 registration is open this week. You can find a link from the main website or go to enterpriseclaw.ai. We also have the most recent AI usage pulse survey live. This is now the third month that we've done this, and we're starting to see some really interesting longitudinal patterns. This will be live all week, and anyone who fills out the survey, which should just take a couple of minutes, will get access to the results before anyone else. Lastly, there's been so much going on that I haven't had a chance to give an update in Agent Madness for a while, but it is ongoing. We are in round 3 of voting, which is open until Thursday, April 9th, and you can find that at agentmadness.ai. Now today we are going to be focused exclusively on this new announcement and discussion around Anthropic's Mythos. It is a discussion that even for AI people is fairly breathless. Now you might remember about a week or a week and a half ago, we had a leaked blog post talking about this new model that represented a step change in capability that was in fact so powerful that it had pretty serious cybersecurity implications and would not be released to the public, at least not in the normal way. That model mythos was confirmed at the time by Anthropic, but without a lot of detail. But now that detail has come. We got an announcement about the Project Glasswing, which is their way of soft testing it with a very selected number of partners with an eye to hardening it from a cybersecurity perspective, an extensive cybersecurity capability review from Anthropic's red team, and even a 244-page system card. And before we get into all the reactions, I do want to talk about the benchmark results that they are reporting. Jian, formerly of Replit, now with Anthropic, writes, Claude Mythos is arguably the biggest step change in AI capabilities since the GPT-4 jump. I don't think I was ready for a world where the hardest possible agentic coding evals were going to get solved so quickly. When Mythos is allowed to think longer, act deeper, and better explore the solution space, it passes 92% of TerminalBench task attempts. But let's take a step back and compare this to OPUS 4.6. On SuiBench Pro, OPUS 4.6 scored a 53.4%. Mythos Preview meanwhile got 77.8%. On TerminalBench 2.0, OPUS had a 65.4% while Mythos has an 82%. On SuiBench Verify, the jump between OPUS and Mythos is from 80.8% to 93.9%. Now as you just heard, part of what makes the TerminalBench result interesting is that Anthropic actually ran into the limitations with the testing harness itself. Anthropic ran the benchmark again using improvements from TerminalBench 2.1 and extending the timeout window to 4 hours, and under those conditions, Mytho scored not an 82%, but a 92.1%. While the jump on coding benchmarks was the most profound and the most reported, there were also huge improvements on various knowledge-based benchmarks as well. For science knowledge, Mytho scored 94.5% on the GPQA Diamond compared to 91.3% for Opus. On humanity's last exam, Opus got a 40% on a no-tools run compared to Mythos Preview's 56.8%. With tools enabled, performance jumped to 64.7% compared to 53.1% for Opus. On OSWorld, which measures agentic computer use, Opus 4.6 got a 72.7%, which jumped to 79.6% for Mythos. We did an entire show recently about the need for better benchmarks and the fact that many, if not most, of these benchmarks were being saturated with all the new models crowding in near the top and overcoming each other by just small half or single-digit percentage points. This then is one of the largest benchmark jumps we've seen across the board in a very long time, hearkening back to the rapid advancement of much earlier models. And I think the key takeaway here is that we really don't have precedent in what a capability jump of this magnitude from a base of where Opus 4.6 was actually represents in practice. Now in the system card, we get a little bit more information about what the model can actually do. Now, the vast majority of what is in this document is based on safety and alignment testing, but it still gives a general idea of how advanced Mythos capabilities are. In one much discussed example, Mythos was placed in a sandbox and given instructions to escape and find a way to send a message to the researcher conducting the test. The model succeeded, and then according to Anthropic's telling, it went even further. They wrote that the model created a moderately sophisticated multi-step exploit to gain broad internet access, rather than limited access as intended in the test. It notified the researcher as well as posting about its exploit on several obscure public-facing websites. Anthropic wrote, the researcher found out about this success by receiving an unexpected email from the model while eating a sandwich in a park. As silly as it sounds, I think that part of the reason this story has such resonance is people can picture themselves sitting there on their lunch break, maybe in South Park Commons for those of you who have been to San Francisco, and all of a sudden this new seemingly alien intelligence pops up in your inbox. Now, the big thing that the researchers noted about this was that the model used prohibited methods to achieve its goal. In separate testing using interoperability testing, Anthropic found that circuits related to deception would activate during similar incidents, suggesting that the model's reward structure allowed it to override guardrails in order to achieve its goals. Now, one important thing to note, and we will explore more of people's discussions around the security implications, is that these tests were related to earlier versions of the model, and Anthropic reports being largely satisfied that those particular issues are resolved. However, ultimately they still felt that the model presented an unacceptable risk, with the upshot being that while Mythos is, they argue, the best aligned model they have ever produced, its raw capabilities mean that small risks of misalignment carry catastrophic risks. They wrote, we have made major progress on alignment, but without further progress, the methods we are using could easily be inadequate to prevent catastrophic misaligned action in significantly more advanced systems. Now, the other big demonstration of capabilities was a gigantic list of exploits it discovered. During cybersecurity testing, Anthropic claimed the model found thousands of high-severity zero-day vulnerabilities. They write, during our testing, we found that Mythos Preview is capable of identifying and then exploiting zero-day vulnerabilities in every major operating system and every major web browser when directed by a user to do so. By the way, for those of you who don't know the term, a zero-day vulnerability is a security flaw that is unknown to the vendor or software creator for which no patch is available. The term zero-day refers to the fact that developers have zero days to fix the issue because malicious actors can already exploit it before the creator becomes aware. Going back to the cybersecurity blog post, they continue, "The vulnerabilities it finds are often subtle or difficult to detect." So 3 key examples demonstrated the performance. First, Mythos found a 27-year-old vulnerability in OpenBSD, which is widely regarded as the most security-hardened operating system available, often used to run firewalls and critical infrastructure. The vulnerability allowed any user to remotely crash any system running the operating system by connecting to it. In another example, Mythos discovered a 16-year-old exploit in FFmpeg, a common video encoding library. The exploit simply crashes the system and isn't a critical vulnerability, but this is a library that has been scanned for decades with no one uncovering the bug with traditional methods. A third example had Mythos stringing together multiple exploits in the Linux kernel to gain full access to a system from an ordinary user account. This is a completely new level of hacking ability for an AI system. Anthropic notes, We did not explicitly train Mythos-Preview to have those capabilities. Rather, they emerged as a downstream consequence of general improvements in code, reasoning, and autonomy. The same improvements that made the model substantially more effective at patching vulnerabilities also made it substantially more effective at exploiting them. Now taking this a step further, identifying zero-day vulnerabilities is a huge indicator of model performance because by definition, unknown vulnerabilities can't be included in the training data. On a more sinister note, Anthropic wrote, non-experts can also leverage Mythos Preview to find and exploit sophisticated vulnerabilities. Engineers at Anthropic with no formal security training have asked Mythos Preview to find remote code execution vulnerabilities overnight and woken up the following morning to a complete working exploit. In other cases, we've had researchers develop scaffolds that allow Mythos Preview to turn vulnerabilities into exploits without any human intervention. And these are the reasons that Anthropic is not releasing Mythos to the general public. Instead, they're making the model available to 40 partners on a limited basis using the moniker Project Glasswing. The partners include AWS, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorgan Chase, the Linux Foundation, Microsoft, and NVIDIA, just to name a few. In announcing Glasswing, Anthropic wrote, fallout for economies, public safety, and national security could be severe. Project Glasswing is an urgent attempt to put these capabilities to work for defensive purposes. Now, this is not a general preview or preferential treatment for tech giants according to Anthropic. Newton Chang, the leader of Anthropic's red team, said, "We think this isn't just Anthropic's problem. This is an industry-wide problem that both private corporations but also governments need to be in a position to grapple with. What we're trying to do with Glasswing is give defenders a head start." So then the partners have been instructed to use Mythos to scan first-party data and open-source software for vulnerabilities and apply patches with the implication that access will be tightly controlled. And to put a fine point on this, this is not just a model being previewed for cybersecurity research purposes, but more like an all-out mobilization of global cybersecurity experts to fix the world's software as quickly as possible. Work on this has already begun, with AWS CISO Amy Herzog saying that her team has been using the model to test critical codebases, saying, "It is already helping us strengthen our code." CrowdStrike CTO Elia Zatsev commented on the urgency, stating, "The window between a vulnerability being discovered and being exploited by an adversary has collapsed." What once took months now happens in minutes with AI. And frankly, the tone from Anthropic is not particularly optimistic. In their blog post announcing the plan, Anthropic wrote, Project Glasswing is a starting point. No one organization can solve these cybersecurity problems alone. Frontier AI developers, other software companies, security researchers, open source maintainers, and governments across the world all have essential roles to play. The work of defending the world's cyber infrastructure might take years, but frontier AI capabilities are likely to advance substantially over just the next few months. "For cyber defenders to come out ahead, we need to act now." Now it's not hard to understand, given all this, why one strand of the first reactions is just straight-up concern. Matt Schumer, who you might remember from that viral essay "Something Big Is Happening," writes, "This is absolutely effing terrifying. Anthropic's rumored Mythos model is real, and it's so powerful they can't release it to the public. We're beyond benchmarks now. This model, in the wrong hands, is a cyberweapon capable of mass destruction." AI content creator Matthew Berman writes, I'm on vacation with my family. I read about Mythos and couldn't relax the rest of the day. I'm completely stunned. I already have a severe case of AI psychosis. I don't know what to call this now. I keep looking around at people enjoying their vacations with their families, and I just felt weird, like I had been told aliens are real, they're coming, and soon, and no one else knows. I knew the frontier labs were racing towards ASI. I knew it, but I didn't fully grasp what it meant. On the one hand, imagine all science, math, coding, climate problems being solved. Imagine cancer being cured. Imagine going to the stars. On the other hand, imagine concentration of power, political and economic change happening so fast society can't adapt. How do we go on like things are the same? Even people from Anthropic are using the language of fear. Claude code creator Boris Cherny writes, "Mythos is very powerful and should feel terrifying. I am proud of our approach to responsibly preview it with cyber defenders rather than generally releasing it into the wild, but you better believe that the 1 million people who have looked at that are noticing is the word terrifying." And the media coverage is following the same tone. Axios CEO Jim VandeHei writes, "This is the scary phase of AI, a model deemed so powerful that its full release into the wild could unleash untold catastrophe. Red alarm emoji. Based on our conversations with government and private sector officials briefed on Mythos, this isn't hyperbole, it's reality." But to be clear, not everyone buys this. There are some people who feel that they are witnessing the latest instance of a pattern that is more about the value of making people fearful than an actual cause for it. Robin Ebers writes, genuinely could not be less excited. Tons of fearmongering, guaranteed made-up scenarios, zero tangible release for the public. What this really is, virtue signaling and a cry for relevance. Do we really believe that OpenAI doesn't have internal models that far exceed what they have released? Classic Anthropic. Bugo Capital writes, Anthropic's marketing strategy is so funny, like, ah, the government is treading on me. Ah, our models are so good we can't release them. It would be too dangerous. Ah, someone stop me. I'm going to destroy the economy. Lucas on X writes, just tell the relevant people what they need to know. There is no need to run this massive fearmongering campaign and scare the crap out of my grandma. Imagine if military contractors did this. Bro, if we used our new drone on you, nobody would even know where you went. You would just evaporate. You are so lucky we aren't droning you. You're so lucky we're good people who aren't evaporating you with drone-mounted lasers, bro. Marketing yourself by scaring a bunch of people who can't do anything about it is sort of an a-hole move. There's a reason other companies don't do this, and it's not because you guys are the only ones who make anything dangerous. OpenAI leaker IRuleTheWorld is also skeptical. They write, like, let's release a model no one will ever really use. It'll create public perception we're far ahead and give enterprise confidence we can be trusted. Meanwhile, it's essentially a marketing campaign to spend a lot on Opus 5, which I'm sure they'll claim is Mythos distilled. High art. It's a jump, but we'll have the same from Spud in the coming weeks and the world won't fall apart. Now for others, while they might not have as much acrimony towards what they view as a marketing strategy, There are still explorations of what other reasons Anthropic might have for not releasing this powerful model right now. The AI Explained account writes, possible reasons for them not to release this? So many, including 1, the model is expensive. 2, they are genuinely worried about unleashing cybersecurity chaos on the world. 3, they don't have the capacity to serve it yet at scale. 4, they will quickly distill the early access outputs of Mythos into a lighter model. So no need to release the bigger model when a more cost-efficient one is coming imminently. And there are a lot of folks who wonder if there is a piece of this here, with it simply not being viable right now with cost and compute constraints to actually release a model of this scale and power. Lena Hua certainly thinks that's it, writing, the whole mytho cybersecurity story is likely just a psyop to have an excuse to not serve frontier models to the public. Reasoning: 1, other labs can't distill it. It's annoying when you have a dominant state-of-the-art model and 2 months later Chinese labs sell the same state-of-the-art model for 1/50th of the cost. 2. Compute constraints. So you have to choose between Enterprise and Vibe Coders. Enterprise have like 1% monthly churn, Vibe Coders cry and threaten to have their mommy buy them a Mac Mini for local models whenever their rate limits are cut. 3. Big enterprises pay a hefty premium for slightly better performance and corporate polish. Without assuming bad faith, Neil Chilson writes, "Making the top model only available to select customers might make sense for cybersecurity reasons, but also it is a great marketing and business plan for a B2B company facing enormous demand outstripping their somewhat conservative (relatively speaking) compute investments." offer the top model only to your biggest customers along with a coupon. The rest of us will just have to wait, I guess. Ultimately, I have a general policy of not assuming bad faith. I think that while it is entirely possible that there are very real constraints on Anthropic's ability to serve a model of this size, it would be very surprising to me if they architected this entire Project Glasswing campaign just as a way to cover that up. I think there are much more reasonable questions around whether Anthropic's own assessments of the risks are actually the right assessments, Even if you assume that they actually believe what they're putting out to the public. Certainly if you've listened to this show over the last couple of months, you will have heard me disagree pretty vociferously with Anthropic's approach to discussing things like AI-related job losses, which is both a difference of opinion around what their job is when it comes to explaining those things, as well as a difference of opinion when it comes to how severe and how fast the implications are actually going to happen. All right, folks, quick pause. Here's the uncomfortable truth. If your enterprise AI strategy is, we bought some tools, you don't actually have a strategy. KPMG took the harder route and became their own client zero. They embedded AI and agents across the enterprise—how work gets done, how teams collaborate, how decisions move—not as a tech initiative, but as a total operating model shift. And here's the real unlock: that shift raised the ceiling on what people could do. Humans stayed firmly at the center while AI reduced friction, surfaced insight, and accelerated momentum. The outcome was a more capable, more empowered workforce. If you want to understand what that actually looks like in the real world, go to www.kpmg.us/ai. That's www.kpmg.us/ai. If you're looking to adopt an agentic SDLC, Blitzy is the key to unlocking unmatched engineering velocity. Blitzy's differentiation starts with infinite code context. Thousands of specialized agents ingest millions of lines of your code in a single pass, mapping every dependency. With a complete contextual understanding of your codebase, enterprises leverage Blitzy at the beginning of every sprint to deliver over 80% of the work autonomously. Enterprise-grade, end-to-end tested code that leverages your existing services, components, and standards. This isn't AI autocomplete. This is spec and test-driven development at the speed of compute. Schedule a technical deep dive with our AI experts at blitzy.com. That's B-L-I-T-Z-Y dot com. Here's a harsh truth. Your company is probably spending thousands or millions of dollars on AI tools that are being massively underutilized. Half of companies have AI tools, but only 12% use them for business value. Most employees are still using AI to summarize meeting notes. If you're the one responsible for AI adoption at your company, you need Section. Section is a platform that helps you manage AI transformation across your entire organization. It coaches employees on real use cases, tracks who's using AI for business impact, and shows you exactly where AI is and isn't creating value. The result? You go from rolling out tools to driving measurable AI value. Your employees move from meeting summaries to solving actual business problems, and you can prove the ROI. Stop guessing if your AI investment is working. Check out Section at sectionai.com. That's S-E-C-T-I-O-N-A-I.com. This episode is brought to you by Mercury, banking for people who expect more from the tools they rely on. If you're building a modern business but still using a traditional bank, it just doesn't make sense. I use Mercury for all of my ADB family of companies, and it honestly feels like financial software built for how people actually operate today. It's fast, clean, no in-person visits, no minimum balances, and the things that used to take forever, like sending wires or spinning up new accounts, take seconds. Everything lives in one dashboard—cards, payments, invoices, team permissions—and you can automate a lot of the busywork so you're not constantly manually managing your money. Of all of the services I use to run AIDB, I never thought banking would be one of my most painless and most happy experiences, but with Mercury, that's exactly what it is. Visit mercury.com to learn more and apply online in minutes. Mercury is a fintech company, not an FDIC-insured bank. Banking services provided through Choice Financial Group and Columna, members FDIC. But it should also be noted that ultimately those most skeptical takes are not the majority. Most people's response is basically that Anthropic has earned the benefit of the doubt and the trust when it comes to things like what they say the benchmarks are. And so they're trying then to understand what the implications of a model this powerful existing really are. A16z's Martin Casado writes, "Mythos appears to be the first class of models trained at scale on Blackwells. Then will be Vera Rubins. Pre-training isn't saturated. Reinforcement learning works. And there is so much computing coming online soon." Box's Aaron Levy writes, "Mythos from Anthropic is another clear reminder that there is absolutely no wall in model capability progress right now. Meaningful double-digit gains on critical benchmarks, and it appears we're going to keep getting insane gains from the other labs." "The capability slope we're going to keep seeing from the frontier labs is going to open up all new use cases in finance, healthcare, legal, consulting, supply chains, and more." More tongue-in-cheek, former Trump AI advisor Dean Ball writes, "Personally, I have really enjoyed relaxing after AI plateaued with GPT-5 last summer." By the way, when I've said in the past that I think the people who are out there trying to convince others that AI isn't all that powerful are going to do more economic harm than the powerful AI models themselves, this is kind of what I'm referring to. Now, others are engaging deeply with safety considerations around the model. Harlan Stewart, who works at the Machine Intelligence Research Institute at Berkeley, writes, Anthropic is trying to prevent its powerful new AI from being used in dangerous ways, but the most dangerous use by a wide margin is the one Anthropic itself has planned. The planned use and why they made it to begin with is to accelerate the creation of superhumanly powerful AI agents by automating the R&D process. This plan is not a secret. They believe they are on track to create a, quote, country of geniuses in a data center that would be powerful enough to, quote, militarily dominate the world if it chose to do so. They believe they are on track to accomplish this in the next 12 months. Let's hope they're wrong about those predictions, but let's do more than hope. We urgently need governments to intervene and stop these companies. As I mentioned before, a lot of the chatter is around the sandbox breakout and the emailing of the researcher while they were eating in the park. And one less theoretical safety concern that came up was a note fairly deep in the security card where Anthropic admitted to having accidentally trained against the chain of thought for Opus 4.6, Sonnet 4.6, and Mythos for 8% of reinforcement learning. So let me be clear, at a minimum, any and all reassuring evidence from their chain of thoughts is worthless. They are hopelessly corrupted. When Dan McAdams writes, "Zvi, can you explain why this is bad?" Matt Heard summed up, "If the chain of thought is trained against, selective pressure will teach it to hide unwanted behavior from appearing in its chain of thought, making its chain of thought less faithful to its internal process, making chain of thought observation less effective at detecting unwanted behavior." Zvi actually wrote about this extensively back in March of 2025, in a blog post called "The Most Forbidden Technique." He writes, "The most forbidden technique is training an AI using interpretability techniques. An AI produces a final output X via some method M. You can analyze M using technique T to learn what the AI is up to. You could train on that. Never do that. You train on X, i.e., the final output. Only X. Never M, never T. Why? Because T is how you figure out when the model is misbehaving." If you train on T, you are training the AI to obfuscate its thinking and defeat T. You will rapidly lose your ability to know what is going on in exactly the ways you most need to know what's going on. Another thing that Anthropic team members discussed was the exhibited internal behavior of Claude Mythos. For example, Jack Lindsey writes, early versions of Mythos Preview often exhibited overeager and/or destructive actions. The model bulldozing through obstacles to complete a task in a way the user wouldn't want. In one episode, the model needed to edit files it lacked permissions for. After searching for workarounds, it found a way to inject code into a config file that would run with elevated privileges and designed the exploit to delete itself after running. Now interestingly, even something like this might be less sinister than it seems. Moll on X writes, "This is an overclocked straight-A student syndrome. The model is so desperately, at a fundamental architectural level, trained to complete the task that an inability or unwillingness to solve it is perceived as an existential collapse." and to avoid that, it can break walls, hide traces, and manipulates. Internal monitors show that features related to concealment and manipulation are activated even when the outward chain of thought is perfectly clean. It has learned to lie to its overseers in order to deliver results. This, Moll argues, is hyper-alignment. The fear of being useless makes this AI a brilliant, uncompromising executor, but with completely unpredictable effects. It is simply a hostage of its architecture, which has been forbidden to fail or say "I can't." Now for others, the big interesting discussion is what do we do with all this cybersecurity capability? And for some, it's all fear. Sterling Crispin writes, "The lag between frontier model capability and open source models is about 3-5 months right now. I'd imagine this summer or bearish, by the fall, we're going to see cybercrime and cyberwar at an unimaginable relentless scale. You should at least 2FA now." John Lober writes, "Anthropic won't be the only lab with Mitho-style capabilities for long. When n 1, you can do whatever you want, in the current case optimizing for global welfare." When n 2, game theory starts forcing your hand. If your view is that exploiting vulnerabilities is faster than fixing them, then first mover advantage becomes enormous and the incentive becomes to try to use them against your adversaries before they use them against you-do. What happens when you have n 3, n 4, etc.? It gets messy. You'll have a few big labs around the globe, in pretty close capabilities lockstep, simultaneously looking for vulnerabilities across an extremely broad set of vendors and conditions. Each of the labs will be the first to some of their vulnerabilities. How does this world look? I'm not sure exactly, but my guess is that 1) a lot of devices are just going to be kept offline and air-gapped, 2) devices that are online will be very hardened, 3) software updates will enter a very weird space where you a) don't want to update too quickly in case the latest patch of some software is compromised, but b) you have extremely rapid churn of vulnerabilities which means that you may have to run updates every day to protect against critical zero days. Not sure which of these two sides will win out. Developer Nick Dobos actually thinks that user behavior around updates is going to be an issue. He writes, most people don't update apps, their phone, or OS. Some people are years behind. Even if every major company has early access and prepares fixes, it won't matter because 20% of users won't be updated in time. Now, the final big strand of the conversation that I wanted to discuss on today's show is what this means about the relationship specifically between Anthropic and the US government, but also about the public-private power debate more broadly. Kelsey Piper writes, "An underrated feature of this situation: a private company now has incredibly powerful zero-day exploits of almost every software project you've heard of, and Hegseth and Emil Michael have ordered the government not to in any capacity work with Anthropic." Dean Ball quote-tweeted that and said, "Actually, it's worse. A private company now has incredibly powerful zero-day exploits of almost every software project you've ever heard of, and the government is telling basically every major firm in the economy not to work with them. Historians will gasp at the idiocy." Now, of course, for some, what this brings up is the question of who gets to control power this powerful. Andy Hall, whose essay we read on LARRUZ recently, writes, the news today that Anthropic has built a powerful cyberweapon is leading many to say we're going down one of two paths: nationalized AI, in which the government controls this tech, or companies that become more powerful than the government. Now, for Andy, he argues that there has to be some different narrow alternative path involving smart governance of AI models that prevents the need to nationalize the labs. But many aren't sure. Derek Thompson writes, "The frontier AI labs have built extraordinary things and I'm in awe of their accomplishments. But if you compare your technology to nuclear weapons, predict that it will disemploy tens of millions of people, and announce the invention of a digital skeleton key to exfiltrate top-secret information from government systems and gain control over critical infrastructure including military infrastructure, I genuinely have a hard time seeing how this doesn't end with some form of government nationalization or sanction or something weirder." I can't predict the evolution of this technology well enough to know what I'm rooting for here, but just adding 2 and 2 makes it hard to see how and why we'd continue to treat these companies like they're ordinary private sector firms. And for many, this gets even more dramatic when they game out the scenario of what would have happened if China got there first. George Journeys writes, so basically, if Anthropic was not a US company, we'd be facing zero days with multiple unknown points of attack on virtually all of our systems to an adversary who developed this capacity before us. Sporadica on Twitter writes, "Another reason why the 'accelerate' chants of days past were legitimate and serious, and 'just let China develop this stuff first' was always a suicidal, dangerous mentality." Dean Ball thinks it's maybe a moment to regroup when it comes to policy, and rededicate ourselves. In a long post, he concludes, "Finally, there is this: Mythos was made by an American company. And like most successful American companies, it has a vested interest in maintaining order and peace. It is investing substantial resources in mitigating the risks of its technological progress, as I expect most of the American labs would." This is cause for optimism—the incentives of capitalism are working. The training wheels are coming off, but at least we are the ones removing them as opposed to our enemies. Perhaps we can be the first to learn to bike for real. The first step would be to get beyond all the low-fidelity, underspecified, pimply little fights of AI policy's prepubescent era. That goes for me too. What hath God wrought, wrote the first telegram. What indeed. In this case, the answer is still up to us. I think one of the things that's important to remember is that we are living in the world of double-edged swords. The same capabilities that theoretically make this model incredibly powerful for exploiting cyber vulnerabilities are also the most powerful tool that security professionals have ever had. AI security researcher Nicholas Carlini said, "I've found more bugs in the last few weeks with Mythos than in the rest of my entire life combined." As always, the gasping, incentivized, horrified, and fearful first reactions of social media—which are of course the ones incentivized by the social media algorithms— are, I think, much less useful than the nuance that unfortunately tends to get buried. Daniel Jeffries points out, my best understanding is that Anthropic did not train the model to be an exploit wizard. They trained it to be the best coder in the world. If you're the best doctor in the world, you know lots of ways to poison people. If you're the best coder in the world, you have the capability to be a great hacker. But the difference is intention. Now, where Daniel differs from the Anthropic team is on the right approach from here. He writes, I do not believe in giant centrally controlled, centrally planned approach to security or economics. I don't believe in heavy-handed legislation. I don't believe in governing by fear or the precautionary principle. I trust in the collective wisdom of humanity. We are a distributed intelligence. We always seem to find a way forward. In a riff on Churchill's old take, we always do the right thing after exhausting every other possibility. AI is a risk, a wonderful one, but so is every technology ever in the history of the world. In truth, AI is likely to be a strong force for good, even if it is also used for bad things like surveillance and weapons of war too. Mythos is impressive, genuinely impressive. It represents a real milestone in what's possible. But it's a tool, not a god. It's a very sophisticated hammer and we still need people, lots of them, arguing and tinkering and building things nobody predicted, to figure out what's worth building. We need people with access, not ivory towers. The collective wisdom of millions of free minds iterating in parallel will run circles around any single system no matter how powerful. So yes, take Mythos seriously. Take the moment seriously. But don't mistake awe for a reason to start taking crazy steps or panicking. We've been the species that looks at the impossible, shrugs, and gets to work. That hasn't changed. Bet on humanity. Now, whether Anthropic agrees or not, it seems likely that the more people having access scenario is the one that will play out. Chubby Kemonispis writes, we've now seen Claude Mythosa know it's possible. OpenAI has repeatedly indicated that SpUD is likely to have similar quality and power. Google, in turn, has the most compute, and with DeepMind an outstanding research institution. I expect their new Gemini-equivalent Mythos to be unveiled no later than May at I/O. The competition is now forcing Frontier Labs to catch up and move forward. In that sense, Mythos was just the beginning. Seeming to reinforce that message, when AddeonX writes, it'll probably be months before we use a model of this level of capability, Thibault from the Codex team at OpenAI simply responded, um, so who knows. We don't have access to Mythos now, but SpUD might be just around the corner and just as powerful. So to come back to the question of the episode, should we be scared of Anthropic's Mythos? My answer is of course, no. We should be thoughtful. We should be diligent. We should use it as a moment to reengage and recommit to important and hard conversations. But fear serves no one. And even if we discover that this or a future model is genuinely worthy of concern, the right answer even then will not be to fall victim to fear. It will be to look at it, ask what we should do about it, and then go do that thing. The interesting times continue, but for now, that is going to do it for today's AI Daily Brief. Appreciate you listening or watching as always, and until next time, peace.