Apologies for being late. When I saw the news about a week ago, I almost choked on my own saliva -- then I immediately rushed to Reddit.
The collective orgasm was so intense that I had to be put in intensive care. Now that I'm out of the ICU, and have recovered my strength, I finally feel ready to upload my thoughts.
***
I know I said I was studying math, but I can't really comment on fluid dynamics and the nature of solutions and so on, so I won't pretend to know enough about the subject (I am still working through fractions) -- I'd be interested in hearing an actual mathematician's perspective on this issue, but so far it doesn't seem like there are any people with such skills involved in the discourse, so we'll just have to wait and see whether the solution is indeed correct.
However the human interest part of the story I *can* comment about, and that is fairly straightforward.
Essentially, a mathematician did all the real work and original thought, but (unbeknownst to him -- let this be a lesson in cyber-security) the chatbot he was using had some gay clause in it or whatever that deprived him of his rights, and he was really being carefully watched and monitored by some retarded nerd. Then the second he was close to being finished, the seething loser nerd just shamelessly stole all the work, had AI compile it, and then said, 'Artificial Intelligence did this, akshually!'.
Then all the paid slopagandist midwits unanimously went, 'See? See? It is *useful* after all! Shut up and accept the data centers! Eat the bugz! Be a good slave! STOP ASKING QUESTIONS! JUST DO WHAT WE SAY!'
Sorry, dork, it doesn't work like that, lol.
They can't even make good propaganda anymore -- now it's just retarded nonsense written by shut-in autistic retards, which is so hopelessly out of touch that it is contradicted by basic observable reality, and therefore totally unbelievable for anybody not part of the echo-chamber.
Anyway, as far as I can tell the only real lesson from all this is to never trust nerds, since they are inveterate power-trippers, and to remember that these freakishly pale useless weirdos only have money and status thanks to the fake bubble tech scam economy. Having no skills of their own, they can only steal from other people, or try to appropriate the achievements of their superior predecessors via appeals to distant genetic lineages and IQ scores.
Now that this is joke of an economy is falling apart, it's over for them, and we at least have something to look forward to.
(Anecdotally, it's revealing how even after many decades of this, you still can't find anybody who views computer nerds as being in the same category as doctors, lawyers, and engineers, in terms of job prestige, respectability, potential marriage value, and so and so forth.
If your daughter was dating a lawyer, doctor, engineer, etc, you'd say, 'Honey, I think you're making a very good choice.'
If she was dating a computer programmer, no matter how rich, you'd
1) Try to dissuade her
2) Tell her, 'I want you to have this. It's called MACE The eyes, aim at the eyes. Remember to never be alone with that creature.'
In fact, the flow of prestige continues to move in the opposite direction. Boys still dream of becoming engineers. Hell, even the nerds themselves constantly try to rebrand their superfluous hobby as 'Akshaully, engineering', whereas no engineer would ever want to be confused with one of them.)
***
I also can't help but notice that the jobs assigned to AI are almost always the most menial, boring, unimportant, rote tasks in the workplace -- it basically never gets used for anything that requires some degree of creativity or inspiration or thinking outside the box.
So, for example, if some aspect of a proof is necessary in order to lay the foundations for the argument, but at the same time is incredibly tedious and lengthy to write (and therefore omitted when the mathematicians are doing original work)... then AI can usually do that just fine, and faster than a human can.
Or if, when writing notes, the researchers abbreviate things / use shorthand / jump around rather than write a fully publishable document... then the AI can fill in the gaps and assemble the pieces together, at least to some extent (meaning: provided there's enough human-generated original material).
Otherwise, it cannot think for itself, and is never used in order to come up with an original solution for a problem.
All this makes it sound more like a glorified secretary or note-taker or low-paid clerk or intern, rather than super-intelligent transhumanist singularity or whatever -- and when you compare the amount of funding that has been wasted on AI, with what it takes to hire a secretary or menial office worker (or H1B, these days), then the whole thing becomes even more ridiculous. It's work you could get from a human for a much lower price!
And all of this is assuming that it can even perform those low-skilled tasks to the same degree of accuracy that a human can...
In reality, anyone who's had to interact with AI knows it often feels like interacting with a clerk who has dementia, and who occasionally makes embarrassing mental lapses you have to watch out for and correct, while otherwise doing decent work. Or a young intern who is too hormonal to care about the job, and so just concocts plausible-sounding, but ultimately meaningless, nonsense in order to skip work -- it lies, hallucinates, confidently gives the wrong answer, tells you the exact opposite of what the data it has in its source material, and will often continue in its wrong-ness no matter how much you prod and push or hint at the right answer.
Everyone has their own version of this when using AI for Google, which is (again) the actual function that it's been reduced to -- notice how steep the downgrade from 'super-intelligent transcendent machine-god' to 'crappier version of Google's broken search engine' is, lol.
And all that, again, at the expense of trillions of dollars -- ridiculous!
Something else they don't say is that, in all these cases, the AI itself is just an inert tool being used by a human user -- the chatbot didn't wake up one day and go 'I'm going to solve Navier-Stokes'. It didn't develop its own without human input. It was just sitting there, with no real life or purpose, until someone (an anti-social power-tripping nerd, in this case) picked it up and used it for something.
ReplyDeleteJust because the machine itself is expensive, complicated, and took a lot of time to design -- doesn't mean anything, or make it less of a tool.
Nobody says, 'the particle collider discovered such and such thing, with some human assistance' -- the actual credit for discoveries goes to scientists, or the lab, or the nation(s) that funded such a project, but not to the machine (again, however useful it may be), which is not autonomous.
Or 'my computer wrote this document, while I pressed the keys on the keyboard', LOL, even if that's technically the case.
Treating AI as if it were human is just schizo slop, akin to talking with your toaster, and ironically risks us descending into believing in magical powers and so on and so forth. Totally unscientific.
Speaking of shameless theft, credit to The Punished Bolshevik Enthusiast on Twitter / X for the memes -- when I do check in on there, it's mostly just to laugh at MAGA. Finally, people are bullying this Thiel-slurping trash, LOL!
ReplyDeleteAnother welcome change, which I hope more people adopt, is to try to use humble self-deprecating humorous usernames, a la 'The Punished' or '3rd worldist' or whatever other variations occur to people.
ReplyDeleteNot in a self-hating way obviously, but more like taking the sting out of whatever insults they're gonna throw at you, by implying that you don't take yourself *that* seriously -- which will actually make you much more relatable and sympathetic toward the audience, compared to the wannabes, posers, and try-hards.
And this also counteracts the sadly widespread trend, especially among paid right-wing slopagandists, of depicting yourself like some ubermensch alpha God or whatever -- aside from the aesthetic cringe and the contrast between their own self-image and their actual status as pathetic sniveling suicidal weirdos, these obsessions are actually a signal of supreme weakness, which is why they're only common among people who lose every fight they get in (as we saw in Yemen, Lebanon, etc).
So being more humble (winners are usually pretty chill) would actually signal success, Darwinian fitness, etc.
I think it somewhat relates to Taleb's example of surgeons that don't look like surgeons -- the grizzled, super-masculine, badass, IN YER FACE BRO, beard-oil / Black Rifle coffee-consuming, warriors in the American army got cucked and routed by the Taliban (who'd never seen the inside of a gym), as well as by the sandle-wearing Houthis.
ReplyDeleteWar is a group contest, not a strength test or other individual metrics assessment, and what it actually tests is cohesion and the ability to cooperate -- and so groups that are made up of LARPers and badasses and other show-boating individualists are in fact actually incapable of cooperation (too self-focused), and thus doomed to just lose pathetically on the actual battlefield, whereas the seemingly weaker-looking groups (on the individual level) are actually composed of humble team-players, which means they are unified enough to actually go on to succeed and win.
So the fact that the online army of right-wingers is composed entirely of Darwin-idolizing LARPing obsessives... is in fact a sign of their tremendous weakness and imminent defeat, LOL (they're just too stupid to interpret this correctly).