frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Expanding Project Glasswing

https://www.anthropic.com/news/expanding-project-glasswing
34•surprisetalk•1h ago

Comments

philipwhiuk•43m ago
It would have been nice to have a list of the 150, but I guess it would make them a hacking target?
frays•42m ago
> We intend to go much further: our longer-term aim is to support the industry in creating new initiatives, standards, and infrastructure for the era of powerful cyber models.

Another reason why I will be buying Anthropic shares after they IPO.

I don't see them as just an AI company, they are a cybersecurity powerhouse too.

Jtarii•26m ago
Thanks for your input Claude.
devmor•11m ago
I see that as not just a spam post, but a generated addition to the dead internet - a real win for us algorithms.
jwpapi•4m ago
Ragebait god
bushido•42m ago
This feels more and more like a marketing/scarcity play for the largest global corps.

Will likely give them time to expand capacity as well. And make them harder to dislodge in these orgs.

mentalgear•34m ago
Here's my big fear: Even IF (and that's a BIG if) we get all critical vulnerabilities fixed in tech (before adversarial/state-actors turn up with open attack models) - we still have (in at least a year) models that will be so good in social engineering that they can still (given enough tokens) gain access to whatever system they want.

If society can't trust banks and other institutions to safely control their data, what follows ?

Do we we collectivelly switch off the internet?

colechristensen•26m ago
Social engineering as a problem goes away when anybody can get a model to do it for them for $5. It stops being possible, it's really the bank's problem when they can't have a minimum wage call center or a robot responsible for people's data.
p-e-w•22m ago
Yes. There will be a few high-profile incidents, and then institutions will be forced to stop performing administrative actions based on people’s word.
applfanboysbgon•8m ago
This outcome is massively detrimental to humanity at large. By eliminating the human factor from support, you make it impossible to get support in edge cases that fall outside of the pre-planned bureacratic process. Everyone already hates that Google can arbitrarily ban anybody they please with no way to get in contact with a human, and you want to extend that to banks in control of people's life savings?
827a•34m ago
GPT-5.5-Cyber has already at least hit if not surpassed Mythos capability in cyber tasks. The only reason they're holding back is because once its out everyone would realize that its capabilities were a step change in March, but are not anymore, yet it costs significantly more and is much slower.
john_strinlai•19m ago
how did you go about assessing this?
jansan•16m ago
So you believe one marketing department more than the other?
NitpickLawyer•8m ago
The brits have a step-based benchmark that they use for this - https://www.aisi.gov.uk/blog/our-evaluation-of-openais-gpt-5...

They seem pretty close, in both average and "best run" scores. And, in a highly verifiable domain, "best run" or best@n is what you're looking for.

cmxch•24m ago
That’s fine as long as I can identify and reject any Mythos derived patch as being irreproducible.
IanCal•9m ago
Why would it not be reproducible?
aspectop•23m ago
i think anthropic is being performative here, creating a hype for mythos and not releasing. i guess this is all a marketing thing to sell a security specialized AI to enterprise and startups at a way larger cost coz security market is deep in money.
jofzar•20m ago
> The organizations in this new group are based in more than 15 countries

I mean most nasdaq tech companies would be in 13+ countries, why are they writing this like it's a big number, is hilariously small?

newtonsmethod•3m ago
I assume they're using a more candid definition where they're not counting all the countries a company may be based, but rather the primary country they're based in.

I don't think they're trying to flex this as a large number. They don't want to give an exact number, as that may change etc / is fuzzy, but also want to give you an idea of the scale.

They say "In the future, we intend to expand our geographical reach much further". I imagine this commentary is somewhat related to the concerns that AI will create an even worse "global underclass". AI developments are first accessible to Americans, then allies, and then later the whole world.

aplthrowaway67•17m ago
How "altruistic" of them. If only Anthropic extended this level of care to the environment or the economy.
tantalor•13m ago
They should share it with Meta.

https://www.0xsid.com/blog/meta-account-takeover-fiasco

fontain•10m ago
“Mythos Preview continues a long-term trend that we’ve been warning about for some time: within 6 to 12 months […]”

The only trend Mythos continues is Anthropic’s trend of warning that disaster is always 6 to 12 months away.

yanis_t•10m ago
Is there any evidence Mythos is qualitatively better than the Opus 4.x?

I'm afraid that the usual mantra that "we just need more scale" that worked well for attracting investments, is not working anymore - bigger models provide marginal improvements while naturally get much more expensive to run.

Is this why both Anthropic and OpenAI are rushing for IPOs this year?

alasano•2m ago
From what I've read so far it's less about Mythos being much better at tasks in isolation.

Security wise, it's about being able to find and chain multiple vulnerabilities to actually create viable exploits.

So I would imagine that if you were using it for regular software development you may not feel that it's that different unless used in a particular way?

Claude Opus 4.8: Capabilities and Reactions

https://thezvi.substack.com/p/claude-opus-48-capabilities-and-reactions
1•7777777phil•39s ago•0 comments

Ableton Extensions (Cdm.link)

https://cdm.link/ableton-extensions-beta/
1•12bits•46s ago•1 comments

Bernie Sanders: A.I. Is a Public Resource. You Should Own Half of It

https://www.nytimes.com/2026/06/01/opinion/artificial-intelligence-bernie-sanders.html
1•surfmike•57s ago•0 comments

The Discovery Problem Is Bigger Than Search

https://sphericalcowconsulting.com/2026/06/02/the-discovery-problem/
1•mooreds•1m ago•0 comments

Sig Node v1.37 KEP Wrangler Interest

https://docs.google.com/forms/d/e/1FAIpQLSeeFtoYLhMGk3Y__YIl50vR_u4Wql9IchwHNnfmHpOvFNg6Qw/viewform
1•mooreds•1m ago•0 comments

Capabilities Can't See Your Agent's Objective

https://jlmr.dev/posts/capabilities-cant-see-your-agents-objective/
1•jelmersnoeck•1m ago•0 comments

SiteGround Faces Backlash After Installing AI Plugin on Customer Sites

https://wp-content.co/siteground-faces-backlash-after-automatically-installing-ai-plugin-on-custo...
1•speckx•2m ago•0 comments

Cloudflare challenge access performance degraded

https://www.cloudflarestatus.com
4•kefabean•2m ago•0 comments

Why Roads Get Washboards – Practical Engineering [video]

https://www.youtube.com/watch?v=3YvwqMDDb3k
1•codeforafrica•3m ago•1 comments

Why 25% of Fuel Stations Could Disappear by 2035, and Which Ones Will Survive

https://driivz.com/blog/gas-stations-closing-by-2035/
1•mooreds•3m ago•0 comments

Pi-session-manager – GUI for browsing Pi coding agent sessions

https://github.com/Dwsy/pi-session-manager
1•flippant•4m ago•0 comments

All GitHub Copilot plans are now on usage-based billing

https://github.com/orgs/community/discussions/197089
2•mittermayr•5m ago•1 comments

Open-source real company briefs to practice AI-native building and get hired

https://github.com/ai-native-builder/ai-portfolio-projects
1•behram0329•6m ago•0 comments

Zork-bench: An LLM reasoning eval based on text adventure games

https://www.lowimpactfruit.com/p/zork-bench-an-llm-reasoning-eval
1•evakhoury•7m ago•0 comments

ECDSA.fail Blows Past Googles Confidential Circuit

https://www.ecdsa.fail/
1•poorman•9m ago•0 comments

Strandfall is now open for sign ups

https://strandfall.com/2026/06/02/strandfall-is-now-open-for-sign-ups/
1•adrianhon•10m ago•1 comments

Japanese home cooking as a household UX

https://washoku.substack.com/p/the-minimum-viable-meal
1•yuki-gohan•12m ago•0 comments

Can your agent delete your production database? [video]

https://www.youtube.com/watch?v=wM4SJsaymUI
1•vnaveen9296•12m ago•0 comments

Preparing for KDE Plasma's Last X11-Supported Release

https://blog.davidedmundson.co.uk/blog/596/
1•jandeboevrie•12m ago•0 comments

Can't find a job after graduation? Blame WFH, not AI

https://www.bloomberg.com/opinion/articles/2026-06-02/can-t-find-a-job-after-graduation-blame-wfh...
3•rustoo•12m ago•0 comments

Show HN: Clipto-fully local, natural language search over terabytes of media

https://www.clipto.com/
1•henry_kang•12m ago•0 comments

The Tedious Power of Storytelling

https://pluralistic.net/2026/06/02/must-we-pretend/
1•hn_acker•12m ago•0 comments

Show HN: Mdrone – a microtonal drone instrument in the browser

https://mdrone.org/
1•gidam•13m ago•0 comments

Are contrastive losses just cross entropy all along?

https://idlemachines.co.uk/essays/contrastive-losses
1•smaddrellmander•13m ago•0 comments

Show HN: Laimen - An Unintrusive Reminder App

https://laimen.app/
1•inventivetalent•13m ago•1 comments

3D-printed book turns its own G-code into raised lettering

https://www.designboom.com/design/3d-printed-book-manual-darius-ou-benson-chong/
1•surprisetalk•13m ago•0 comments

Checking Assembly with Z3

https://bernsteinbear.com/blog/asm-z3/
2•ibobev•14m ago•0 comments

Is the Monaco Grand Prix decided at qualifying?

https://entropicthoughts.com/is-monaco-decided-at-qualifying
1•ibobev•14m ago•0 comments

Screamer: Fast Subnet Discovery Without Scanning Every Host

https://github.com/ifritnoises/screamer
1•ifritnoises•14m ago•0 comments

Partial Pathfinding Applications

https://buttondown.com/jaffray/archive/partial-pathfinding-applications/
1•ibobev•14m ago•0 comments