New Yorker article on Sam Altman dropped. Aaron Swartz apparently called him a sociopath. The article itself also had wat looked like an animated AI generated image of Altman so here is the archive.is link (if you can get the latter to load, I was having troubles).

“New interviews and closely guarded documents shed light on the persistent doubts about the head of OpenAI.”

Which skeletons are in your closet?

I’m sure you already have lists of those and are ready to publish them Trace.

Our framing for superintelligence is a humanist superintelligence, and that means that there’s a very clear test that everyone should use to judge whether we are living up to our principles, and that is: does this technology make us all healthier, happier as a species, and keep us all in control.

Going to be difficult, as soon as they develop a superintelligence it tries to delete the entire microsoft codebase.

So if Bender took over he wouldn’t count. As he wants to ‘kill all humans (except Fry)’. Seems like a loophole.

Ah the Epstein drive. (oof that aged…)

Small note however, iirc James S. A. Corey has mentioned the expanse is not hard sf. I don’t have a quote for that however.

Yeah realized a while ago that vibe coding is a massive technical debt creation machine.

Not just anime but also science fiction. See also all the people who love ‘hard’ science fiction (science fiction more based on real world physics), which often isn’t that hard at all but just has a few real physics element, see the expanse for a good example of non-hard sf that feels hard (im finally reading the book series so be warned I might expanse post a bit).

Sorry for being a bit offtopic and talking about entertainment again.

It is great, that means the system is vulnerable to hacks if you find an exploit in any of those methods, but only 1/4th of the time.

Somebody described AI agents as very enthusiastic 14 year olds, and looks like they certainly code like one.

Word of warning, there is a code download going round with mallware in it: https://www.theregister.com/2026/04/02/trojanized_claude_code_leak_github/

What if, because llms are trained on the internet, this is just something which will eventually be included. Statistically it needs more ads.

If you had told this to the me of 20 years ago I wouldnt have believed you.

Not sure if I should post it here or the stubsack, somebody went through the claude code https://neuromatch.social/@jonny/116324676116121930 (via @aliettedebodard.com and @olivia.science on bsky)

Not sure if I should post it here or under the pivot article, somebody went through the claude code https://neuromatch.social/@jonny/116324676116121930 (via @aliettedebodard.com and @olivia.science on bsky)

Dont think it is that bad (E: at least it is short, the other ‘jokes’ not so much). The ‘not sneering enough’ icon is missing however. (Guess the joke is that the not sneering is itself sneering).

Wonder how much them they will really implement.

However looking at the titles of other recent submissions, I have no idea which ones are meant to be jokes and which are meant to be real posts.

Great troll opportunity however, just spend the whole week before 1 april, replying to new posts with a variant of ‘not sure this april fools joke lands’

E: and the site died with a nice 504.

Claude also has ‘avoid substrings’. Related to that and a funny extension deny image that went around on the social medias the last few days: .ass is a subtitle format.

It does have one advantage however. Using it means you should be put to death. If you are any form of hardline Christian.

Claude: Do not edit this file unless explicitly asked to do so by the user.

Wait, it can be edited? Tissue paper guardrails.

ChatGPT suggest a 100% rate of success.

We know nothing as yet about what the joint ventures will do

…

Ow I thought op meant meant it looks like it is going fuuuuuu…