News for hackers. Stuff that matters.
Re:Project Glasswing: what Mythos showed us by roxolotl on Monday May 18, 2026 @12:58PM (#48182252)
What does this mean?

> It's a different kind of tool doing a different kind of work, and that makes a clean apples-to-apples comparison to earlier models difficult.

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.

Re:Project Glasswing: what Mythos showed us by JeremyNT on Monday May 18, 2026 @04:01PM (#48184785)
> 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.

Re:Project Glasswing: what Mythos showed us by password4321 on Monday May 18, 2026 @03:29PM (#48184355)
> 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.

Re:Project Glasswing: what Mythos showed us by __natty__ on Monday May 18, 2026 @01:16PM (#48182473)
Sounds different because it’s hidden advertisement not a regular blog post
Re:Project Glasswing: what Mythos showed us by grim_io on Monday May 18, 2026 @01:20PM (#48182511)
But why would cloudflare advertise Anthropic? They are competing with Anthropic by hosting open weights models.
Re:Project Glasswing: what Mythos showed us by Someone1234 on Monday May 18, 2026 @01:26PM (#48182601)
Re:Project Glasswing: what Mythos showed us by hdndjsbbs on Monday May 18, 2026 @01:38PM (#48182761)
It's circular financing. It's a circlejerk. It's a circular financing jerk.
Re:Project Glasswing: what Mythos showed us by neuronexmachina on Monday May 18, 2026 @11:38PM (#48188912)
Can you elaborate? That just seems to be a Cloudflare's announcement from May 2025 that they'd be supporting MCP servers.
Re:Project Glasswing: what Mythos showed us by drak0n1c on Monday May 18, 2026 @07:27PM (#48187292)
They got privileged early access to an unreleased frontier model to harden their systems, with Anthropic engineer support, and likely were able to use it to make other product optimizations tangential to security too. A blog article afterwards is a cheap price for unlocking that access, regardless of how well it paid off.
Re:Project Glasswing: what Mythos showed us by fmbb on Tuesday May 19, 2026 @02:23AM (#48189900)
A corporation does not have its own will.

Their owners are invested in AI and need AI to do well. If this goal clashes temporarily with the goals set up for Cloudflare, so be it.

Re:Project Glasswing: what Mythos showed us by perching_aix on Tuesday May 19, 2026 @02:48AM (#48190073)
Not how I took it, but that this was a lot more marketing than content, while their other, older blogposts are more content than marketing. It is of course all content marketing, for Cloudflare. That doesn't mean it has to be bad (anemic on meaningful content). On the contrary, it being good is kind of the whole point.
Re:Project Glasswing: what Mythos showed us by alsetmusic on Tuesday May 19, 2026 @10:09AM (#48193540)
> Sounds different because it’s hidden advertisement not a regular blog post

Yep. Cloudflare has lost my respect over the last six months.

The posts about pro-AI initiatives and APIs for AI and then laying off a lot of people was pretty impressive for how to do the wrong thing.

Re:Project Glasswing: what Mythos showed us by eikenberry on Monday May 18, 2026 @02:45PM (#48183773)
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.
Re:Project Glasswing: what Mythos showed us by rs_rs_rs_rs_rs on Monday May 18, 2026 @03:19PM (#48184204)
It is not, that's what surprised Anthropic employees too.
Re:Project Glasswing: what Mythos showed us by smusamashah on Monday May 18, 2026 @03:29PM (#48184358)
'Its not X, its Y' is also a common LLM trope.
Re:Project Glasswing: what Mythos showed us by FergusArgyll on Monday May 18, 2026 @01:20PM (#48182509)
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

Re:Project Glasswing: what Mythos showed us by Xirdus on Monday May 18, 2026 @01:37PM (#48182740)
But did they build this different harness? And are they sure other models can't cope with it?
Re:Project Glasswing: what Mythos showed us by roxolotl on Monday May 18, 2026 @01:50PM (#48182932)
Right I expected the piece to transition into “and here’s how we built a whole new thing for it” but it never did.
Re:Project Glasswing: what Mythos showed us by jascha_eng on Monday May 18, 2026 @03:00PM (#48184002)
They kind of have a little diagram explaining the steps I imagine every single step in that to basically be it's own Claude code session.
Re:Project Glasswing: what Mythos showed us by meander_water on Monday May 18, 2026 @05:25PM (#48185886)
> 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

Re:Project Glasswing: what Mythos showed us by _alternator_ on Monday May 18, 2026 @06:40PM (#48186795)
There's pretty strong evidence that (mis)alignment in one area creates (mis)alignment in others. The "aligned behavior" vectors are not orthogonal from cybersecurity to bioweapons to prejudice, so having alignment in some will likely bleed into others.
Re:Project Glasswing: what Mythos showed us by sn0rl27 on Tuesday May 19, 2026 @02:14PM (#48197061)
The model wasn’t created specifically for security research. It’s a general model that just happens to be dangerously good at security research (according to Anthropic)
Re:Project Glasswing: what Mythos showed us by samstokes on Monday May 18, 2026 @05:43PM (#48186115)
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.

Re:Project Glasswing: what Mythos showed us by getnormality on Monday May 18, 2026 @08:17PM (#48187685)
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.

Re:Project Glasswing: what Mythos showed us by mycall on Monday May 18, 2026 @11:12PM (#48188757)
Google Maps is my favorite human harness.