Fable 5 is the same underlying model as Mythos 5, but with cybersecurity and biology blocks. Mythos is the first model that's made me feel that we've entered the next phase of model progress. For years, we've talked about cybersecurity / self-improvement / autonomy / model-dominated coding / biology implications of model progress. Some of these are issues to defend against; some are areas to advance. Mythos has made me & our team feel like we've seen the earliest glimpse of the world we've been talking about. Also, we published a lot of cyber eval results in the system card, including some evals we designed recently, as well as details of safeguards. In most cases, Mythos 5 ~= Mythos Preview. We found it ticked up on the new ExploitBench eval, and we opted to put that in the eval table so people can calibrate/update on advances in cyber capabilities to be prepared for. (We don't want to compete on offensive capabilities and don't try to.) But overall, Mythos 5 is an efficient model, about equal to Mythos Preview in most cases. I'd really like more people to design new security evals! The better models get, the more our limited evals only see a small part of the picture. In terms of where we go from here, here are some current thoughts: 1/ It's important we get Mythos cyber capabilities to defenders. We just have to do it safely and cautiously. We're working on an expanded trusted access program. We're working with government and industry to do this. I sort of envision the next 1-2 years being a large scale effort to make the world resilient design & implement new approaches to security. 2/ I think cybersecurity will start merging with AI security and alignment. Let's say you're a defender and you want to use a model -- will it break out of its sandbox? Will it stop where you tell it to stop? This is one reason I'm excited about working on cybersecurity. In the limit, it's the same thing as AI security. 3/ I really want people to develop new evals for... defensive cybersecurity, hardware security, autonomously running a business, advanced biology, and other parts of national security. Our internal eval ship rate is way, way up because Mythos makes it easy to iterate, especially on the engineering aspect of building evals. (Sometimes, we ask new hires to make a new eval on their first day, and another on the next). I’m excited we’re making this available as Fable 5, because I think the world spending time with the model is the most important way to calibrate.

This is good! I started red teaming LLMs for biorisks/weapons risks in my bedroom in Nov 2022. In 2023 a lot of people said we were overreacting -- 'models won't be better than internet search'. True then, but the important point is they were going to get a lot better. Now I think there's clear reason for caution. The ultimate solution is inventing deploying biodefenses. This is one step. I fully expect humanity to conquer pandemics in the 21st century

A lot of people have been wondering about Mythos, Glasswing, and the vulns we / our partners are fixing. Today, I’m excited for us to start sharing more. (For context, I lead Glasswing @AnthropicAI.) Two independent evaluations this week—from XBOW and the UK AISI—confirm what we've been seeing internally: Claude Mythos Preview is a step change in autonomous cybersecurity capabilities. We need to start preparing fast for a world of models with this level of capabilities. The UK AI Security Institute tested the model we shipped at the launch of Project Glasswing and found Mythos Preview is the first model to solve both of their end-to-end cyber ranges, including one (Cooling Tower) which no model had ever cleared. But attackers (and defenders) have sophistication & cost constraints – Mythos is also the only model that clears every one of their tasks estimated over 8 hours under their deliberately low 2.5M-token cap. XBOW tested it on their offensive security benchmarks, finding "token-for-token, unprecedented precision." It's the only model to succeed at subtle V8 sandbox work. Other Glasswing partners shared similar stories. In a few weeks of testing, Mythos Preview has helped them find many thousands of (estimated) high critical severity vulnerabilities, sometimes double what they'd normally find in a year. I don't share this to boost Mythos. In fact, this is not about Mythos. It’s about preparing for the coming world of models being better, faster, cheaper, and more creative than some of the best human experts at dual use capabilities. Clearly, we need them supporting defenders as widely as can be done safely – and especially the least resourced ones. Within a year, Mythos will probably look quite dumb (relative to other new models). And others may release openly available or unguardrailed models of Mythos-level capabilities. We started Project Glasswing because capabilities like Mythos Preview's won't stay rare, or stay in careful hands. We are bringing it to defenders as fast as we responsibly can, while working to figure out, for example, the right safeguards and patching & disclosure processes. Also, to be clear, compute has never been a limiter in our rollout. Expect a fuller update on our Glasswing work in the coming days. XBOW report: xbow.com/blog/mythos-offensi… UK AISI report: aisi.gov.uk/blog/how-fast-is…

I'm optimistic this eventually favors defense over offense. We wanted to start this transition cautiously. I've honestly been inspired by what orgs have been able to do with Mythos. More to come!

The team I lead is part of The Anthropic Institute. The way I think about the Institute is "applied weird blue sky research". So far we've got a good track record of tackling some of the most important ideas -- cyber, self-improvement, robots, national security.

Also, if you're a security researcher / leader really motivated by the mission of "solve the whole AI cyber problem", you should apply to Anthropic. We're looking e.g. for vulnerability researchers, senior security researchers and engineers, AI security research leaders, etc.

Seeing this on Slack that day was one of the first "oh, I guess we're just seeing it now" moments for those who think about AI security