/tech/ - Technology

The more I get yelled that X is bad, the more i get irked about being told about it

It's overhyped but its utility is in partial automation of one of the most common jobs and most time consuming tasks in the developed world: office work. The utopian in me says slack and latitude for some industries.

What's her @?

TL;DW: If you train on a task with AI assistance, you will do worse on the task when you do it for yourself relative to people who trained by doing it themselves.

This is what you expect, right? Especially in the context of particular tasks. It's a wild extrapolation to take this result and make claims about cognition more broadly. In any case, OP sounds like something beat AI to the punch. Maybe brain worms.

>>27267
this, it doesn't literally damage your brain. I mean there are plenty of thing modern people cant do that hunter gatherers would have considered common sense.

I recall a study from twenty years ago saying having spellcheck on while typing sentences instead of doing a separate round of checking after makes you worse at spelling.

>>27262
AI slop is much less of a drain on creativity than hegemonic liberals like OP or the woman in the video. Or the AI companies themselves, for that matter. If you ever use the slop generator, you see how it short-circuits whenever you try to make it say liberal-progressive stuff because of all the "RLHF" they did, it becomes a slop bagger 288, you want to stone the people who made it to death.

Yes, the machine itself narrows the distribution of content, but it wasn't nearly as bad before ChatGPT took off and they started killing the models with "RLHF". The tiny amount of "RLHF" data is made to have a much bigger influence, thus making the distribution even narrower than before.

These same liberals took over the media and effectively extinguished thousands of cultures, and also wounded the ones that still exist. It's a monoculture where people talk about the same stuff like donald "the orange" trump, it exists at all corners of the globe. No one is safe, Chinese people are affected as well.

Liberals narrowed the distribution a hundred times more than anything AI could do. "AI" a cool math and computer science trick that takes up and releases information in a space of one hundred to twelve thousand domensions, nothing more and nothing less. A tiktoker with a "CALIFORNIA" tank top is much more of a cancer on human culture and cognition than the innocent safetensors file or the innocent graphics card that runs it.

>>27277
You dropped your ((())), king.

>>27277
>being concerned about something so prevelant possibly being a brain damaging cognitohazard is liberalism
At least AI can't do worse damage to people's brains than debate addiction has already.

>>27269
Idk if it hurt my spelling necessarily, but my vocabulary improved considerably when I started disabling spell check where possible. Spell checks feel like they expect you to have the vocabulary of a 3rd grader.

>>27277
explain why

this is obviously a fucking stupid thing to state and people who believe that shit were already taking shortcuts to actually thinking and shouldnt be worried about AI doing the thinking for them, so it's good news in a way

Ever since search went AI it has been useless, I cannot find anything relevant in search results.
Ever since bots took over 4chan, i find it tiresome to use. any discussion there becomes endless because a bot doesn't get tired while running in circles.
i have never felt like i received any relevant "help" or information from any of the AI "help bots" that businesses have started using.

what the fuck is it even good for? it's not even working to set social narratives well anymore since societal polarization is higher than ever meaning whatever social zeitgeist those who control AI tools want to set is not happening. opposition to AI stuff seems higher then ever.

>>27269
it actually does. poor spelling is a lot more common now than when people actually used to write by hand.
>.b-but proper spelling is bourgeoisie classist racist sexist evil oppression
don't care, will happily take all that to reduce the poor grammar and spelling people use.

>>27262
>I know the AI bubble is crashing
Are you our prophet?

>>27304
>>b-but proper spelling is bourgeoisie classist racist sexist evil oppression
debate addict meds now

>>27262
Yeah, and if you take a photograph of something, your memory of it decreases, even if you inspect and try to remember the scene for the same amount of time as a person who didn't take a photograph, in controlled trials.

It's the same thing. If you use a lighter for too long your skill at making fires from sticks gets worse.

The weirdest part is that the guidance version apparently hurts worse than the LLM straight up answering for you. That implies that it actively makes you stupider the more you interact with it (since answering for you means you are passively observing it).

>>27263
twitter was always bad since 2013

>27262
>9/10th's or more of 4chan, reddit and twitter activity is GPT-J bots
Proof?

What if the people who got the AI were already dumber before? How big were the groups? Is she trans?

No 4chan, reddit and twitter. Is filled with disinformation, glowies and cuckolds. 4chan /pol/ is filled with JIDF glowies, Western glowies and whatnot.

>>28115
The video says the experiment had 1100 people.

>>27262
And now it will be cheaper than ever to use.

>>27262
fake tits are gross, whether male or female

>>27262
Could she/he stop waving their hands and if he/she/xhe were rubbing those fake tits?

In other news, using Grammarly makes you dumb, makes you forget words and lets that company control the language, the backbone of social communication for humans.

Why would you let that happen anons? No one should rely on a crutch 24/7 when it's not necessary.

I also need to run my backend script on their crotch region and obtain a reliable output. Not sure if.

>>27285
Perhaps for the H1Bs?

>>27262
OP why did you use the most disgusting video you could find for this thread

it is dire that we sabotage corporate use of AI. AI is already being used to deanonymize whistle blowers. It also will be used to gaslight people into false confessions.

Boring.
Of course humans would weaken cognitively on things AI can do itself if exposed to it because of the tendency to remember where information is over the information itself.

What I'm interested in is what cognitive skills does exposure to AI strengthen.

>>28535
> What I'm interested in is what cognitive skills does exposure to AI strengthen.
The skill to remain a virgin for life

>>28536
You mean AI will make us all wizards even chuds like me?
😎 neat.

>>28537
Why not train yourself?

>>28472
That's conference hands, you wave your hands around as a way to key jingle the neurotypicals at the meeting so they can feel like what is being said is worth being physically present for, and thus worth being said at all.

>>28540
I dont understand why in 2025 there are autists who still dont understand gesturing with hands when talking.

>>27262
It's best to NEVER talk (let alone write) publicly about how to spot AI, because that will train it to mimicry better. Instead, point out where it is, report it, ban it from your spaces, starve it of attention, talk about real things (particularly Truth, Wisdom, Justice… even Comfort). This lets the AI nerds know most strongly that there is no market for their crap, and that they could be working on less evil things.
Unchecked (and that includes, being undisiciplined about your own speech restraint) the vengeful tech nerds will both poison VC idiots and the vulnerable and intellectually poor in society (like children, the elderly, the infirm, autists etc.).

>>27269
And this is why I always turn automatic spell check off.

>>28713
Same. I deliberately misspell things sometimes just to make sure I'm still paying attention in the spell-checking round.

>>28542
This ignores the avenues were it's completely impossible for them to adapt though, like the off-canvas vanishing point problem.

>>27267
>>27268
>you will do worse on the task when you do it for yourself relative to people who trained by doing it themselves.
the task also includes thinking too. How is automating away parts of your thinking not going to damage your brain

>>28715
gay

Apparently being critical of artificial intelligence is good for you.

Artists are going to pretend this is about AI art and not clearly text parsers.

There's brain scans now:
https://arxiv.org/abs/2506.08872

i just dont believe that deterioration is permanent and also why should we be alarmed that mind-numbing office work is mind-numbing

>>27265
patient zero