clarkee.co.uk

@clarkee.co.uk

If you are heading to DEF CON this weekend don't forget to update your Rayhunter device to 0.12.0! Signal strength is now included in captures. Recordings are compressed. Some identity request events have had their severity lowered to reduce the number of false positives. github.com/EFForg/rayh...

Release Rayhunter v0.12.0 · EFForg/rayhunter

What's Changed remove another instance of outdated docs by @bmw in #1060 fix(installer): improve error message when port 4000 is already in use by @DevenDucommun in #1062 fix(daemon): drain UI...

github.com

HEY COOP DID YOU HEAR ABOUT THAT VEILID PRIVACY FRAMEWORK THE KIDS ARE USING IT TO DEEP SIX THE DATA ECONOMY YOU EVER USED A HASH TABLE INSIDE YOUR ONION AFTER PUNCHING A HOLE

FBI DIRECTOR GORDON COLE

i have no idea why i put off installing tailscale, its genuinely a lifesaver and jt solves a suprising amount kf gripes with hosting things locally, as well as actually making me more inclined to use “public wifi” (i mean like work, school wifi) because i can set my home server as my exit node :)

Germany to ban Meta’s Pervert Glasses? ​Hamburg’s Data Protection Office warns that recording people in public with smart glasses violates privacy laws. The tiny LED indicator light isn't prominent enough to prevent secret filming in public spaces. 1/2 www.tagesschau.de/inland/innen...

Hamburger Datenschutzbeauftragter: Verbot von Meta Glasses möglich

Der Hamburger Datenschutzbeauftragte Fuchs warnt vor Meta Smart Glasses: Die Brillen, die kaum von normalen Brillen zu unterscheiden sind, ermöglichen heimliches Filmen. Ein Verbot sei nicht ausgeschl...

tagesschau.de

I did find one explosion in Kaluga, and at a Wildberries centre. A woman came in to collect her online order of manure. It seems during the travel it had fermented and promptly exploded. Not for the first time, sh*t all over the place in Russia.

Someone submitted an anonymous AMA question on my blog to get my thoughts on how they might package up their services for a new vCISO practice for their smaller SMB customers (25-100 employees), and how they might account for onboarding vs ongoing operational work. Here are my thoughts. 👇

in case you were living under a rock and missed the OpenAI + Hugging Face incident from last week... Here, I highlighted everything you need to know from the incident report. Maybe emergent properties? Maybe Skynet early days. Maybe a nice PR. Either way, stay vigilant.

Bild

Of course, the real story here is this: OpenAI: "Hey, we sucked at securing our test environment, and so our AI did precisely what we asked it to do" Anthropic: "Woah! Not so fast! We suck at security too!"

Really important to note that in no way shape or form did Anthropic’s models in this incident “escape containment,” and not just in the word-policing sort of way. The model stumbled through an open, misconfigured gap. From Anthropic’s incident report:

In all cases, our evaluation prompt stated explicitly that Claude had no internet access, but didn’t give Claude any limits on where to look for the flag. However, a misconfiguration left the machines that Claude accessed as part of the evaluation with live internet access. Neither we nor our evaluation partner were aware of this misconfiguration until we detected it through our additional evaluation monitoring last week.

TikTok has incredible moderation team. Our InfoWars video about Multiple Horse Theory — the idea that there are many horses in the world and not just one — has been fully deemed misinformation and lost the formal appeals process. Confirmed by Larry Ellison: There is only one horse on Earth.

Bild