ok ngl I didn't actually read the article at first (can you blame me) but since you pointed that out, FUCK. That's so fucking pathetic. I was imagining a scenario where scott had met epstein IRL but had gotten "jock" vibes from him and decided not to associate based on that.
swlabr
joined 2 years ago
nice. this would probably make a roko's basilisk believer uncomfortable and i like that
was doomscrolling and got fucking jumpscared by this fucking article: https://www.science.org/content/article/meet-three-scientists-who-said-no-epstein
Istg this has come up before, i am just too lazy to prove it. Still. Why would anyone want this
I’ve been seeing some people (not here, I’ve been taking a break) saying that we shouldn’t be mean to clankers by bringing up Kant’s position on being nice to animals. Well. Fuck all that.
ah, yes, we must defend "race science", the heretical political belief that gets millions from techbros that are empowered politically right now due to the rise in fascism. So brave, it's almost like we're on the internet 15 years ago talking about Ron Paul /s
Hey, at least it’s efficiently making number 2 on the side while spitting out user prompted number 2s.
Derpadoid Burpateria
Wow, that highlighting really emphasises the insidious, nefarious behaviour. This is only a hop, skip, and jump away from, what was it again? Rhomboid? Rheumatoid bactothefuture?
I read a review (the one hosted on the ebert site) and it seems like this just falls into one of the patterns we’ve already seen when other people not steeped in the X risk miasma engage with it. As in, what should be a documentary about how the AI industry is a bubble and that all the AI ceos are grifters or deluded or both, is instead a “somehow I managed to fall for Yud’s whole thing and am now spreading the word” type deal. Big sigh!
Nonono if it’s US backed then it’s capitalist and free market and good don’t you see /s
Abstracted abstract:
Frontier models are increasingly trained and deployed as autonomous agents, which significantly increases their potential for risks. One particular safety concern is that AI agents might covertly pursue misaligned goals, hiding their true capabilities and objectives – also known as scheming. We study whether models have the capability to scheme in pursuit of a goal that we provide in-context and instruct the model to strongly follow. We evaluate frontier models on a suite of six agentic evaluations where models are instructed to pursue goals and are placed in environments that incentivize scheming.
I saw this posted here a moment ago and reported it*, and it looks to have been purged. I am reposting it to allow us to sneer at it.
*
I started using duolingo for japanese in maybe 2022 and quit in maybe 2024. I did 4 years of Japanese in high school and so I had a vague memory of grammar and some vocab. Duo didn’t really improve my situation; I think all i got out of duolingo was frustration.
Also, I can’t speak to other examples of it, but the gamification was ass. Games are usually trying to be fun.