Linxz

Idk man VR and shit.

[Linxz @LinxzSec]

@fwrnr @JeffWellsRigInt 🙈

[Linxz @LinxzSec]

@udunadan Most bugs are garbage and writing real world usable exploits for them is still a giant wall.

[Linxz @LinxzSec]

It’s so not over. We are so back.

[Jeremiah Grossman @jeremiahg]

Project Glasswing:
* 37,966 findings.
* 2 exploited in the wild.

https://t.co/IRH3iMaltM

[Linxz @LinxzSec]

@udunadan In the back of my mind I knew all along the LLMs never had the makings of a varsity athlete.

[Harry Stabbings @harrystabbings]

NEW: Households will have to register their Wi-Fi password with Ofcom under new Online Safety Act guidance.

Guests must show ID before connecting.

[Linxz @LinxzSec]

ID to listen to music!? We are so past it mate.

[Politics UK @PolitlcsUK]

🚨 NEW: Under-16s in the UK could now be banned from Spotify, Lego Play and YouTube Kids under expanded Government proposals to widen the scope of its social media ban

[Waterworld Capital @WaterworldCapi1]

What's the charge? Getting Honeypotted? By a Succulent Chinese Spy?

[Linxz @LinxzSec]

I’ve recently switched off Cursor (because of the changes to subscription) and the codebase indexing feature is so underrated.

GPT is absolutely retarded without it.

[pam @pamoutaf]

As promised, here is the blogpost for code execution on the meta rayban 😎 https://t.co/m1GWZiOCF2

[Linxz @LinxzSec]

Took quite a frustrating situation for me to internalize this so clearly.

[Linxz @LinxzSec]

Successful research is often about identifying the correct areas to put attention. LLMs are such a force multiplier here because you can put some attention into more areas, while also having it largely not be your own.

[Linxz @LinxzSec]

@UndedInside Not sure yet.

https://t.co/Re9RW1yjkv

[Dave Aitel @daveaitel]

Y'all. A good Linux kernel exploit writer might find 30 bugs in their whole career. But: https://t.co/gg1oDGxJNc ????

[Linxz @LinxzSec]

It’s fucking over.

[Linxz @LinxzSec]

@hexadeclan 🤔

[Linxz @LinxzSec]

ifykyk

[Linxz @LinxzSec]

The defense rests, your honor.

[Sean Heelan @seanhn]

haha, so apparently OpenAI are about to start training on your Daybreak Blue sessions, with no way to opt out.

Literally stealing people's private work, as they do it, is cartoonishly evil.

It also will presumably kill Daybreak usage by security researchers. Or at least, it should. The thing OAI have had to swear they are not doing to math researchers, they are now explicitly about to start doing to security researchers.

Yet another reminder of the importance of open models.

[Linxz @LinxzSec]

@XploitBengineer Has he considered merging with Anthropic? Sounds like if they pooled their GPUs it would be better for everyone. /s

[Linxz @LinxzSec]

That’s certainly a take…

Not like they hyper regulate who has access to cyber capable models, what they can use them for and how much they vacuum up or anything…

[morgan — @morqon]

sam: local inference is a terrible allocation of scarce resources

“a lot of people want local inference, and somehow they assume that the problem is solved now because we have open-source models, we're going to be able to run them on our computers, and so we're not going to need all the stuff in the data centre

but from a global allocation perspective, actually it's quite terrible to have everybody, like, have a bunch of the world's limited supply of DRAM sitting in their own computer and being used five percent of the time

you want maximum utilisation in the cloud if you're trying to deal with this very supply-constrained environment

somehow many people's intuition is the opposite of that”

[“paula” @paularambles]

what is the charge? running a model? a succulent chinese model?

[Linxz @LinxzSec]

@roddux @gabsmashh I am feeling very Tony Soprano today.