Ce qui me tue c’est qu’un vrai gros bras ne se serait pas laissé encercler comme ça. A force d’impunité les flics sont devenus gras et fragiles, ils se sentent tellement en sécurité qu’ils ont perdu toute conscience tactique de la situation autour d’eux.
C’est une information importante pour les mois qui viennent parcequ’il y a tout un tas de méthodes collectives qui permettent de nier une asymétrie d’armement et de moyens.
As with everything, the poison is in the dosage. All of your examples are categorically different from what’s happening here.
You make a game with deep mechanics, and there is some fucked up shit that can be done in it, accidentally or on purpose ? Who cares. You make a game that is uniquely about torture ? Yeah you’re a weirdo and i’m keeping tabs on you if i can. You make something that isn’t even gamified and is just a virtual torture dungeon ? You’re 100% a psycho, normal people don’t do that.
Same with movies. Saw is a tame product of Hollywood professionals who decided to push the cursor a little further than is usual. Love it or hate it it’s just a movie. Now if we’re talking real devious shit like human centipede or Poughkeepsie tapes etc… Those are products that barely have a story, and are mostly sadistic exhibitionism. Some guy (who sometimes isn’t even a film maker) has this repulsive idea he wants to force on people, gets their family and friends to pitch in a bit of money, then writes, directs, edits, and markets the thing solo over years of his life. Not something normal people do.
My wife is, among other things, a horror writer. So naturally i agree there’s nothing wrong with exploring morbid ideas and pushing the envelope. But some people do that with no fun, no game, no art to it, no statement, just the obsession of unlimited sadistic control. And it’s weird, not ha-ha anti-conformist weird, just “cover your drink” weird.
Yeah I think people are missing the forest for the trees. There’s no reason to worry for the AIs but using your free time to conceive a virtual torture dungeon is not normal. Totally agree with your point, this is a symptom of a society that normalizes not just violence but sadism too.
Yes, good music sits outside of the statistical distribution of existing music. It is mathematically impossible (or at least astronomically improbable) that a model could ever stumble on it.
The tragedy of my life is that I used to put a lot of dashes in my comments - not em dashes, just plain old minus signs. They’re not really used in my native language but I picked that up from reading Edgar Poe when I was a kid.
Oh no I don’t think we would. Once an asteroid is earth bound and past its apoapsis there is very little you can do (even with near infinite propulsion) to make it not collide with the earth.
Destroying it turns one big problem into a billion small problems with the same velocity and trajectory, which is also a no win scenario.
Freedom of expression is not a complex issue. If you’re the top dog on a tour and some promoter kicks one of the acts out for just expressing an opinion on stage… It seems simple enough to stand up for your colleagues.
Your fascist hate group is a total fabrication. Sheeran is not a victim of anything, he just chose to side with the rich and powerful over his music “friends”, which will never get you any sympathy points from the general public.
I would imagine that this is a good way to get rid of the initial employees who really believed they were doing good. Now if you stay after this, or if you get hired after this, they can be confident that you have no moral issue with what they are doing.
squishy is a better term than “hard”
Eh, agree to disagree. My use of the phrase might be easier to parse if you think about the “hard problem of consciousness” as opposed to the “easy problem of consciousness”. Also, more remotely, the notion of NP-hard vs NP-easy in computational complexity. I guess we are arguing about semantics aren’t we ?
I’m fairly certain that’s not a strict definition - LLMs also seem very heavily influencable via their context window inputs, though those are more limited and transient than the training set.
While LLMs don’t simply “statistically predict the next token” (common over-simplification which bears no relation to what is actually going on), the output they produce does exhibit the same statistical distribution as their training data. That’s how you get intelligible language, code that compiles, chains of thought that makes sense etc… The context input will steer the output towards a certain subset of the corpus (highly optimized C functions have slightly different distribution than throw-away Python scripts), but it can never direct the model fully out of distribution as those out of distribution vectors cannot be expressed within the model’s embedding space.
she was a commodity - as talented as you could ask, more talented than many national acts, but not unique
Yes exactly, if music was solvable then the most skilled people would systematically outclass the least skilled ones which is obviously not the case.
it’s precisely because software engineering is hard (not difficult, but rigid - objective) that makes it a good fit for LLM agent execution
Yes i think we’re actually in agreement here. I said “hard” (not difficult) as a reference to “hard problems”, a term that comes from complexity theory but is now commonly used to describe problems which can’t be reduced to an algorithm or evaluated objectively, and thus can’t readily be “solved”.
Squishy subjects like music and sociology are full of hard problems, while solid subjects like math and coding are full of easy problems. Now the change introduced by LLMs is that as long as a problem is “easy”, it can no longer be so laborious as to be impossible. Every software problem is solvable, modulo the effort/computing power you can spend on it.
Your job as a software architect / engineer is actually to find the easiest workable solution(s).
You also get bonus point if your solution is average (standard, unsurprising etc…), which makes it particularly soluble in LLMs which, by definition, can only produce output that is within the distribution of their training set.
As for relative difficulty, the “soft” fields have a very sliding scale for that with a lot of allowance given to newbies that isn’t accepted “at the highest levels”
That’s not where i would put the difference. If you take a field like music, the problem is that it can’t “just work”. A nostalgic song may move the masses today but you can’t say “okay we’ve solved nostalgia let’s get to serenity next”. Soon enough you’ll need a new nostalgic song and by definition it will be out of distribution. You can’t find it in a high dimensional representation of past music, and, well, you can’t train on future data, so there is no way an LLM finds it and recognizes it for what it is.
with a high bar of standards required before publication and release, LLMs are a powerful tool that can accelerate many creative processes
I still believe they’ll never amount to much regarding artistic processes, and not just for the reasons i already mentioned. To make something good you need to sit with it and walk with it and spend some time in it doing all the tedious little tasks until it really feels like home and you can express yourself in it. You can’t achieve that if a machine speedruns all the little tasks for you.
You’re entirely right. This won’t replicate to other fields like writing and creative arts in general, but software engineering is just not that hard and can basically be brute forced with a good harness.
It’s a done deal and there is no world where people will write professional code by hand. I like it cause it really separates coding (the job) from coding (the art form). People will code by hand for aesthetic reasons just like people learn the violin instead of using a synth and we’ll have a generation of lovingly crafted stuff. But boring software will be entirely automated, if not generated on the fly based on immediate needs.
That’s cool but you’re describing the models from 2 years ago and also not considering harnesses, which account for most of the progress of the last year or so.
The whole thing is predicated upon him training her to sit still for hours on end on stream “like a prop”, which she barely ever does. Litterally you just have to watch the VODs , you’d be lucky to find more than 10 minutes of her on-screen on most streams. The whole scenario is pure brainrot and only believable if you don’t look critically at the evidence. So this guy used a shock collar to train a large dog to sit still for hours on end but there was only one (1) transgression ever recorded and she never sits still anyway. How do you square that up ?
You’re grasping at a few seconds of video (out of the 6+ hours he produces daily year in year out), edited out of context by people with a very specific agenda, and walls of concern-trolling prose that tugs at your emotional strings for some reason. It’s trivially easy to make up that kind of story and astro-turf it to death, with hundreds of inactive accounts coincidentally posting their first comment in the same week, which happens to be a copy-pasted list of pre-digested arguments and fact-free appeals to outrage. You’re getting QAnon’d in the worst way.
people engaged enough to believe a wall of text on Reddit or a video compilation on YouTube, but not engaged enough to actually watch his channel to check anything
Hey it’s cool man, i sure expected to see one of you people in the replies.
Same, not that I hate them specifically but I have a skill to completely ignore artists I don’t care about. Bunch of my friends were big fans but me I can’t even tell them apart from Linkin park. Like, who sang “numb” ?
I discovered him when they tried to pretend he was zapping his dog on stream. It was an interesting sequence because apparently they targeted people engaged enough to believe a wall of text on Reddit or a video compilation on YouTube, but not engaged enough to actually watch his channel to check anything.
Obviously they found a large audience, but hey Hasan gained a follower I guess.
Probably grants, and even when exercised they almost certainly don’t come with voting rights.
The book burning thing is just a public relations campaign by publishers because they are getting ready for another billion dollar lawsuit and they want the public’s approval.
There is no shortage of books we print like 10 billions of them every year, and the publishers and booksellers burn a lot more books themselves than AI labs ever will.
Insolite : les détenteurs de l’autorité publique mentent sans vergogne pour justifier leur campagne de violence