Hacker News

> They claim it’s a different kind of tool and then describe using it the same way you’d use any other model. This really felt way worse than the average Cloudflare blog and really just rehashed the Mythos announcement which had already called out the key parts being chaining and crafting examples.

Hah, I was trying to parse this too.

Charitably perhaps they're being vague on exactly what's different because they're still under NDA.

> way worse than the average Cloudflare blog

How long has it been since you took your average? Lately all Cloudflare output has been heavily AI'd.

I think they're saying it has qualitatively different capabilities that make certain kinds of security work more worth pursuing with the model, not that the model of human-AI interaction has changed.

You're right that they're using a harness like everyone else. The general idea of giving the model a harness is not going to change. I mean even humans need harnesses to accomplish some things.

Sounds different because it’s hidden advertisement not a regular blog post

> the model has its own emergent guardrails that sometimes cause it to push back on legitimate security research requests. But as we found, these organic refusals aren’t consistent - the same task, framed differently or presented in a different context, could produce completely different outcomes as illustrated in the examples below.

This was new. I'm surprised that a model specifically designed for security research and gated to professionals is refusing legitimate requests

The post says they wrote a custom harness that orchestrates work between multiple separate model invocations. That is different from running Claude Code (which is a specific existing harness around the Claude models).

The post takes a while to get around to saying that, and could have included more detail besides the workflow diagram and table (which they flag as only "an example of" such a harness), but it does answer the question. It's a different kind of tool because it's a model rather than a harness+model pair.

'Its not X, its Y' is also a common LLM trope.

My guess is because it is a model trained specifically for security/hacking. So comparing it to Opus, trained for chat/code/etc., is apples-to-oranges.

I think what they might mean is:

Because of it's capabilities, a new kind of harness can be built for it, thus the entire system (model + harness) is a different kind of tool than say Claude code

I'm still waiting something more specific or groundbreaking too. Feels like a lot of noise with just the goal to get people to talk about it. And now I realize I am talking about it and about nothing at the same time. Just fugazzi.

I think the idea here is you give the Hunters (stage 2) a narrower scope, but have a parent agent responsible for dividing up the full search space (stage 1).

And note that Hunt tasks can be queued from previous Trace tasks, ie you find a vuln in one layer, so you queue a hunt for corresponding vulns in the layers that could exploit your first finding.

Yeah this whole post reads like Anthropic said “make sure you say how awesome Mythos is” but really what they’re saying is that it’s just a better harness.

Who is him?

In the last few days I was recommending to read the insights from XBOW [1], it's a competitor but it adds more information to the discussion.

[1] https://xbow.com/blog/mythos-offensive-security-xbow-evaluat...

Sentence constructions like this definitely scream AI: "That's a reasonable bias for an exploratory tool. It's a ruinous one for a triage queue..."

I will upgrade the "why it matters" to "and now AI output is part of the training data". A day is coming when the punched-up AI verbiage will be the norm and hard to distinguish unless you're from the previous generation. Sort of in the way that I miss some aspects of Usenet.

It's fascinating seeing people think that if you're snarky enough about something, the substance of that thing actually ceases to be substantive.

It's like staring down the barrel of a gun and taking the time to make quips about the type of paper the gun advertisement was printed on.

This is not just any large organization, it's Anthropic. Their entire shtick is that AIs can do Real Work now and it'd be weird if they didn't behave accordingly themselves.

This is also why Claude Code is full of weird bugs and why their support says that it did refunds when it didn't and so on and so forth.

Cloudflare blogs have been excellent for many years, long before transformers arrived.

This looks more like it was edited by AI rather than fully written by it. Or they are using a really good humaniser for the second pass.

Should that be surprising? Larger orgs are the ones more naturally associated with mediocrity and are most likely to want to reduce human labor hours.

Disappointing really.

I don't know, but it always seems weird to me when people notice AI isn't performing super well and then they conclude that the solution to problem is to try using more AI

Or they don’t, and they* sell access to Mythos and successors through their services company or network of partners and charge a premium.

* they, I mean all foundation models providers, as OpenAI seems to go in the same direction

As much as I’d like to share in the skepticism, the very beginning of the article states it very plainly — this is a step function.

Lots of people feel that Mythos is a psyops campaign, but I don’t really understand the skepticism. Most of it seems to stem from the general distrust of things that aren’t publicly available.

A few Anthropic employees have described Mythos as a general purpose model improvement, but that claim has yet to be widely backed up so that’s the only place I’m remaining skeptical.

For the domain of security research, I’m willing to buy the narrative.

They specifically describe that exploits are usually multiple small vulnerabilities chained together. With that understanding, it sounds like closing vulnerabilities isn't the same as discovering an exploit. Instead, you're leaving fewer small gaps behind, to make it harder & harder to put together a working exploit.

I've settled in on the opinion that it's much more creative and able to run agentically for longer periods of time. So, despite it not having drastically better "hard skills", it's able to combine those together in more effective ways.

Right now, many of these vulns are identifiable by Opus, but they still require a human-in-the-loop (and often a skilled one) to guide towards complex exploits. Without a human in the loop, this means it's a lot easy for the average person to identify and leverage an exploit.

Palo Alto Networks released patches for their firewalls for a number of CVEs last week, almost all derived from their access to frontier models including Mythos.

https://security.paloaltonetworks.com

Most of their new products are AI tools that nobody uses, so I guess they’ll keep posting slop. And recently, they’ve fired so many people that they probably don’t have good writers anymore.

Yeah I’m waiting for this as well.

I get that you want to address them or whatever before releasing info but I keep seeing these claims with barely any data and I’m like…how do you expect people to not be skeptical?

I mean hell if you’re a security professional you’re literally paid to be skeptical.

The part on the harness is spot on.

I have been encouraging people to think about agentic coding in the same way.

Let agents do the reading and writing and inspections. Human does the thinking.

Asking an agent that is looking at a firearm specification schematic "what is wrong with this?" and the response is "this thing contains an explosion and can kill". Human "that's the function" when the human should be asking "based upon the materials used, are the fault tolerances sufficient to maintain structural integrity".

even careful programmers working in unsafe languages will introduce bugs; it's inevitable. in 2026 we should be using safe languages for all new projects, but there's a gargantuan amount of C/C++ handling protocols.

but I agree that guardrails will only help for like, 3-6 months. we should be screening as much as we can with Mythos; unfortunately, Anthropic is only giving access to the big players.

2 things can be true at the same time.

I think the curl folks finding it underwhelming is more of a testament to their code being subjected to a lot of tests/attacks/auditing over the past years compared to many other codebases. It's not going to find magically insurmounable exploits on it's own and "pwn teh w0rld".

At the same time, there is so much shitty non-memory safe code out there (C/C++ mainly) or logically weak code (much of it vibe-coded or otherwise by inexperienced devs) that will be easy pickings for anyone pointing Mythos at those codebases/services and eventually lead to chaos since the cost of an customized exploit has gone from days to months of expensive researcher time to some token spending.

Now if they noticed that they could find exploit chains easily in a lot of popular software, some embargo and hardening to give popular OSS packages time to not be exploitable by default does help people (and the NSA that probably has a preview).