Happy #Caturday!

#Cativitypub #feliverse #CatsOfMastodon #tortitude

I wrote a few words about a brilliant investigation by Laura Shin, who asked a suspected North Korean spy during a job interview: "Can you say something negative about Kim Jong Un?" It's a surprisingly effective tactic that works even today as companies face the growing threat from fake North Korean IT workers.

https://this.weekinsecurity.com/this-quick-question-can-still-expose-north-korean-spies-in-an-instant/

This fall, Apple will be cutting off millions of iPads from future OS updates. The solution seems obvious: Let us install a different operating system.

This is already an option for Macs, but it's not for iPads. Couple this with how difficult these tablets are to repair and you've got a big problem for people who want to keep their iPads running past their expiration date. We think there's room for improvement. Read more at the link below.

https://www.ifixit.com/News/118176/apple-could-give-obsolete-ipads-new-life-with-this-one-weird-trick

Replace Ads with They Live style slogans

A fork of uBlock Origin Lite that, instead of hiding cosmetically-blocked ads, replaces them with white tiles bearing slogans from John Carpenter's 1988 film They Live: OBEY, CONSUME, WATCH TV, SLEEP, SUBMIT, CONFORM, STAY ASLEEP, BUY, WORK, NO INDEPENDENT THOUGHT, DO NOT QUESTION AUTHORITY.

(Use the original @ublockorigin for safety!)

👁️‍🗨️ https://github.com/davmlaw/they_live_adblocker

#addblocker #AddBlock #theylive #ublock #movies #scify #ublockorigin #ads #slogan #WebAds

"The ppl of the West are faced with a confusing juxtaposition. Nearly everyone agrees about what architecture is beautiful and which cities are nice ... Yet the world keeps churning out architecture that is not beautiful and cities that are not nice." https://www.bensouthwood.co.uk/p/eight-architecture-myths-busted-by | nitter
https://nitter.net/DouthatNYT/status/2087222355739230697#m

#architecture #beauty

Idly musing that the Therac-25 cautionary tale I learned about in my undergrad software engineering course resulted in a total of (IIRC) 6 medical injuries over about 4 years. And that was considered horrific and required immediate changes to how computers were used in medicine.

saw this on BlueSky and instantly thought @futzle would be interested ... some fascinating Oddnesses in here, mostly that kilo and milli are both a 1000 measure, but we don't use them that way now ...

It’s odd seeing different decades of #computing #history compressed into one “not now” era, these days. People talking about Novell and IPX over an image of a Cisco VoIP phone, or an image of a PC with a 50x CD writer next to a turbo button and a MHz display. We’ll be seeing punch cards next to a Sinclair Spectrum, next.

Nothing really mattress.

Reverse Centaur, the book by Cory Doctorow, digs into the weird math of AI financing. It gives an idea of the scale of the scam: “a shitpile this big has to have a pony inside somewhere!” There will never be safe autonomous cars, flawless code, or new medicine. There is no growth. The scam is real though. Get angry. #fuckai

This article starts with a story from someone who tried Claude Code and found it amazing, but then switches to the same person a few months later seeing what a disaster it’s been. This quote is key:

The problem is that code produced by an AI agent looks reasonable, but can contain ‘hard-to-spot bugs’ that end up causing major problems

LLMs, by their nature, generate statistically plausible output. That is often a set that overlaps with correct output. But the things that are not correct look exactly the same as the ones they are.

Learning to review code is hard. You look for the bugs that you expect to be possible implementing it by using your theory of mind for the person writing the code and the kinds of things that they might overlook (not necessarily a specific person, but the kinds of things people miss) and also common bug classes.

And the big help is that the person writing the code is not thinking adversarially. They are not trying to sneak bugs in. Normally. Unless they’re a supply-chain attacker, and we’ve a depressing amount of evidence that code review doesn’t catch supply-chain attacks.

An LLM is not trying to do anything. It has no intent. But it is a machine that is trained on code that made it past code review. The kind of bugs that it will generate are ones that look like code that appeared in production. This is exactly what an attacker would do: try to write code that looks correct but is subtly wrong.

I’m only being slightly flippant when I say LLMs are a mechanism for bringing supply chain attacks in house.

Yes, you could pick up a guitar, learn some scales, go behind the woodshed and practice until your fingers bleed, buy a 4-track, record a crappy demo, and repeat until you’re good, like some sort of caveman, but why bother when you can pay some money, type a prompt, pay some more money, and get “your” creation delivered to you, on authentic vinyl with real artwork, ready to display next to your Harry Styles records and the old Madonna and Pearl Jam albums you inherited from your parents?

Doris
#WhiskersWednesday #feliverse #Cativitypub #CatsOfMastodon #tortitude

Musical slop merchants Suno have figured out how to bestow Authenticity on AI-generated music: by pressing it to vinyl:

https://themusic.com.au/industry/nightmares-on-wax-suno-launches-new-vinyl-pressing-servce-for-ai-generated-music/hfQ7mZibmp0/12-08-26

If enough techbro money gets put into this, getting non-slop music pressed to vinyl may go the way of the consumer PC market.

Am I the only person who can hear vaporwavey TV-ident background music made with late-80s Roland digital synths when they look at these screenshots? @nina_kali_nina https://tech.lgbt/@nina_kali_nina/111857639206801638

one more fun fact about cats:

in South Korea, cat lovers are called 집사 or jibsa, which can be translated as “butler”.

we’re cat butlers. feels accurate!

> Our lab has agreed on a blanket ban on the use of generative AI for writing.

reason:

> Using generative AI inherently exposes you to the risk of career-ending accusations of plagiarism.

https://www.carlsonlab.bio/thoughts/the-only-reason-youll-ever-need-not-to-write-with-ai

#TIL the director of The Brutalist was the guy who played the “UFO abductee” in Gregg Araki’s “Mysterious Skin”:
https://www.indiewire.com/features/interviews/brady-corbet-mysterious-skin-interview-1235209735/

You're right - AI _is_ useful. Your use of it provides an important signal to me:

1) For writing: "I didn't care to put the effort in to write this myself."

2) For coding: "I don't care if I understand the solution."

3) For vuln impact analysis: "I don't understand the code base nor the attack vector."

Or shorter: "I do not care about the details."

This, of course, on top of the other ethical problems with AI. It's a strong signal, and I see it everywhere.

Inkscape developers need help to push version 1.5 across the finish line. If you have C++ experience and hate Adobe, consider getting involved

https://jlai.lu/post/41763129