Small models also found the vulnerabilities that Mythos found • Hacker News Board

Re: Small models also found the vulnerabilities that Mythos found

#171

Post by apgwoz » Sat, Apr 11, 2026, 8:58 PM UTC

BoiledCabbage wrote: post #150

integralid wrote: post #98

Earlier quoted context omitted.

>Or none We already know this is not true, because small models found the same vulnerability.

> because small models found the same vulnerability. With a ton of extra support. Note this key passage: >We isolated the vulnerable svc_rpc_gss_validate function, provided architectural context (that it handles network-parsed RPC credentials, that oa_length comes from the packet), and asked eight models to assess it for security vulnerabilities. Yeah it can find a needle in a haystack without false positives, if you…

The benefit here is reducing the time to find vulnerabilities; faster than humans, right? So if you can rig a harness for each function in the system, by first finding where it’s used, its expected input, etc, and doing that for all functions, does it discover vulnerabilities faster than humans?

Doesn’t matter that they isolated one thing. It matters that the context they provided was discoverable by the model.

Re: Small models also found the vulnerabilities that Mythos found

#172

Post by slopinthebag » Sat, Apr 11, 2026, 9:09 PM UTC

antirez wrote: post #16

Congrats: completely broken methodology, with a big conflict of interest. Giving specific bug hints, with an isolated function that is suspected to have bugs, is not the same task, NOR (crucially) is a task you can decompose the bigger task into. It is basically impossible to segment code in pieces, provide pieces to smaller models, and expect them to find all the bugs GPT 5.4 or other large models can find. Second:…

Idk, it seems reasonable to me

> "Our tests gave models the vulnerable function directly, often with contextual hints. A real autonomous discovery pipeline starts from a full codebase with no hints. The models' performance here is an upper bound on what they'd achieve in a fully autonomous scan. That said, a well-designed scaffold naturally produces this kind of scoped context through its targeting and iterative prompting stages, which is exactly what both AISLE's and Anthropic's systems do."

Also they included a test with a false positive, the small models got it right and Opus got it wrong. So this paper shows with the right approach and harness these smaller models can produce the same results. Thats awesome!

So, if you're struggling to make these smaller models work it's almost certainly an issue of holding them wrong. They require a different approach/harness since they are less capable of working with a vague prompt and have a smaller context, but incredibly powerful when wielded by someone who knows how to use them. And since they are so fast and cheap, you can use them in ways that are not feasible with the larger, slower, more expensive models. But you have to know how to use them, it requires skill unlike just lazily prompting Claude Code, however the results can be far better. If you aren't integrating them in your workflow you're ngmi imo :) This will be the next big trend, especially as they continue to improve relative to SOTA which is running into compute limitations.

Re: Small models also found the vulnerabilities that Mythos found

#173

Post by nfcampos » Sat, Apr 11, 2026, 9:11 PM UTC

Anthropic marketing (and even supposedly technical write ups) sadly has become more hyperbole and less substance over time imo. This technology is so impressive on its own, really feels like shootings themselves in the foot in the long run, but what do I know

Case in point here where they conveniently fail to report the false positive rate, while also saying that if it wasn’t for Address Sanitizer discarding all the false positives this system would have been next to useless

Re: Small models also found the vulnerabilities that Mythos found

#174

Post by mattmanser » Sat, Apr 11, 2026, 9:11 PM UTC

WhyNotHugo wrote: post #142

johnfn wrote: post #15

The Anthropic writeup addresses this explicitly: > This was the most critical vulnerability we discovered in OpenBSD with Mythos Preview after a thousand runs through our scaffold. Across a thousand runs through our scaffold, the total cost was under $20,000 and found several dozen more findings. While the specific run that found the bug above cost under $50, that number only makes sense with full hindsight. Like any…

OTOH, this article goes too far the opposite extreme: > We isolated the vulnerable svc_rpc_gss_validate function, provided architectural context (that it handles network-parsed RPC credentials, that oa_length comes from the packet), and asked eight models to assess it for security vulnerabilities. To follow your analogy, they pointed to the exact room where the gold was hidden, and their model found it. But finding t…

Or would it have any way if they hadn't pointed it at it? Who knows?

Just like people paid by big tobacco found no link to cancer in cigarettes, researchers paid for by AI companies find amazing results for AI.

Their job literally depends on them finding Mythos to be good, we can't trust a single word they say.

Re: Small models also found the vulnerabilities that Mythos found

#175

Post by decidu0us9034 » Sat, Apr 11, 2026, 9:12 PM UTC

ang_cire wrote: post #86

SCHiM wrote: post #48

Earlier quoted context omitted.

You can imagine a pipeline that looks at individual source files or functions. And first "extracts" what is going on. You ask the model: - "Is the code doing arithmetic in this file/function?" - "Is the code allocating and freeing memory in this file/function?" - "Is the code the code doing X/Y/Z? etc etc" For each question, you design the follow-up vulnerability searchers. For a function you see doing arithmetic, yo…

That's funny, this is how I've been doing security testing in my code for a while now, minus the 'taint analysis'. Who knew I was ahead of the game. :P In all seriousness though, it scares me that a lot of security-focused people seemingly haven't learned how LLMs work best for this stuff already. You should always be breaking your code down into testable chunks, with sets of directions about how to chunk them and wh…

In a large codebase there will still be bugs in how these components interoperate with each other, bugs involving complex chaining of api logic or a temporal element. These are the kind of bugs fuzzers generally struggle at finding. I would be a little freaked out if LLMs started to get good at finding these. Everything I've seen so far seems similar to fuzzer finds.

Re: Small models also found the vulnerabilities that Mythos found

#176

Post by john_minsk » Sat, Apr 11, 2026, 9:15 PM UTC

johnfn wrote: post #87

kilpikaarna wrote: post #83

Earlier quoted context omitted.

Wasn't the scaffolding for the Mythos run basically a line of bash that loops through every file of the codebase and prompts the model to find vulnerabilities in it? That sounds pretty close to "any gold there?" to me, only automated. Have Anthropic actually said anything about the amount of false positives Mythos turned up? FWIW, I saw some talk on Xitter (so grain of salt) about people replicating their result with…

> Wasn't the scaffolding for the Mythos run basically a line of bash that loops through every file of the codebase and prompts the model to find vulnerabilities in it? That sounds pretty close to "any gold there?" to me, only automated. But the entire value is that it can be automated. If you try to automate a small model to look for vulnerabilities over 10,000 files, it's going to say there are 9,500 vulns. Or none.…

In the future there shouldn't be any bugs. I'm not paying $20 per month to get non-secure code base from AGI.

Re: Small models also found the vulnerabilities that Mythos found

#177

Post by youre-wrong3 » Sat, Apr 11, 2026, 9:18 PM UTC

johnfn wrote: post #87

kilpikaarna wrote: post #83

Earlier quoted context omitted.

Wasn't the scaffolding for the Mythos run basically a line of bash that loops through every file of the codebase and prompts the model to find vulnerabilities in it? That sounds pretty close to "any gold there?" to me, only automated. Have Anthropic actually said anything about the amount of false positives Mythos turned up? FWIW, I saw some talk on Xitter (so grain of salt) about people replicating their result with…

> Wasn't the scaffolding for the Mythos run basically a line of bash that loops through every file of the codebase and prompts the model to find vulnerabilities in it? That sounds pretty close to "any gold there?" to me, only automated. But the entire value is that it can be automated. If you try to automate a small model to look for vulnerabilities over 10,000 files, it's going to say there are 9,500 vulns. Or none.…

[dead]

Re: Small models also found the vulnerabilities that Mythos found

#178

Post by palashdeb » Sat, Apr 11, 2026, 9:18 PM UTC

Been tracking this since the blog post, quick a big deal they are making it.

Re: Small models also found the vulnerabilities that Mythos found

#179

Post by decidu0us9034 » Sat, Apr 11, 2026, 9:28 PM UTC

nfcampos wrote: post #173

Anthropic marketing (and even supposedly technical write ups) sadly has become more hyperbole and less substance over time imo. This technology is so impressive on its own, really feels like shootings themselves in the foot in the long run, but what do I know Case in point here where they conveniently fail to report the false positive rate, while also saying that if it wasn’t for Address Sanitizer discarding all the…

Right now, we accept false positives as long as you can sort them out. I think it's pretty typical that >99% of fuzzer runs don't result in new coverage. Of course they're far from useless without feedback but it's better to have it if you can. I guess the question is does the llm approach have lower costs for validation and triaging vs just fuzzing alone, unclear to me. Anthropic would like people to believe automation is this scary new unknown

Re: Small models also found the vulnerabilities that Mythos found

#180

Post by boelboel » Sat, Apr 11, 2026, 9:28 PM UTC

slibhb wrote: post #124

The best way to think of Anthropic's communication about Mythos is as advertisement. It's basically "our model is too smart to release" which suggests they're ahead of OpenAI (without proof)

Seen similar things with Openai and Palantir.