‘equivalent to being around one year of schooling ahead of their peers’
https://www.youtube.com/watch?v=RcRY6M5_mvo&list=UU9rJrMVgcXTfa8xuMnbhAEA
https://pivottoai.libsyn.com/20260908-students-who-dont-use-ai-are-ahead-on-every-measure - podcast
time: 4 min 14 sec
To anyone who would like to read the primary source, the corresponding part of the OECD’s official report starts at p. 236.
Disclaimer: I have not watched the video in the OP, so I can’t make any statement about whether or not the report really aligns with the video.
I expect we’re gonna hear more about the cherry-picked data points that aren’t atrocious for AI. Like specifically the “help me learn” usage stat (the only one where some AI users are on par or better), and the even more specific “help me learn” usage with AI/media literacy discussed in class stat.
There’s probably all kinds of interesting factors involved that had to be ignored because they needed an AI fluff statistic. Like the one fact they mention, that high AI-use and partly non-AI use correllate with curiousity. Kids who want information will use the tools given to them. If your teachers are into AI and discuss it in class, even more so. I expect the “once or twice a year”-users (who bottom out every chart) to include a big bulk of kids that just do not care about education (often coming from unpriviliged households), so they use it to cheat on their assignments and call it a day.
Pretty undeniable though, that the daily users are always worse than the sometimes users. That’s probably the habitual delegation of thought kicking in. I could accept that a subgroup of the AI users learns on par or sliiiightly better than non-users on average, but they might be building up a reliance on tooling that will serve them worse and worse once they get past well-trodden school problems. In the same vein they’re probably all getting fucked, because noone gives a fuck about good text books & reference materials anymore.
maybe this is the wrong thread to not read the article and just sit there and goon about how the thing with empirically bad outcomes is “just a tool”
There should be an internet-wide bot that replies to every “just a tool” defence with “you’re just a tool”
Now for the big question:
Did AI make these students stupid, or is it just selection bias because the stupid students are more likely to use AI?
I suspect it’s a bit of both.
We have plenty of evidence that states that practice is an excellent learning tool. We also know that AI removes the need for practising thinking by providing immediate answers for everything.
We also have evidence of students making the poor decision to cheat instead of learning, so it’s logical to assume the same type of student would turn to AI.
They can skip doing their homework without the immediate consequences of skipping their homework.
I also think it’s both.
The ones who were already stupid are more likely to use AI, because that’s a stupid decision.
But some students were probably fine, but using the AI atrophied their thinking skills.
Thats cool as shit, I didn’t expect the results to be so obvious for a few more years.
‘AI’, as Artificial Intelligence, is yet to be demonstrated, and the opinion of anyone convinced it has needs to be summarily dismissed, and if they persist, ridiculed.
LLMs are a chipper and solvent of competency and credibility, which is a problem when used by someone attempting to build and develop them.
Are you talking AI or AGI, aka. “Hollywood AI”?
Either, intelligence requires possession and presentation of comprehension, which LLMs are intrinsically incapable of, and I’m yet to observe any example of otherwise.
So you’re talking AGI, or “aware/comprehending” AI, which is a thought experiment at best.
We’ve had good old AI since sometime in the 70s (60s?). It’s used everywhere, though mostly in navigation and control.
In the past 20 or so years computers have become powerful enough to do Deep Learning (aka. it would take a human more than a lifetime to understand the models), which has resulted in CNN, LLM, Generative AI, and all those other glorified Random Number Generators.
AI, contrary to AGI, has nothing to do with comprehension. It’s basically just a big lookup table, either created by humans or by running some machine learning on a dataset.
See also the “Genetic Algorithm” and “K-Nearest-Neighbor”, some of the first MLAI implementations, and simple enough to run on a Windows 98 computer.
Hes saying that calling it “intelligence” is a misnomer and misrepresents its actuality, I think y’all are talking past eachother
His point is rooted in the semantics of the term “artificial intelligence”
You’re proving my point, jog on.
And you’re proving mine, as in the majority of people not knowing history and simply gobbling up whatever those tech bros serve you.
you seem pleasant
nobody asked you to poorly explain some shit we already know
now jog the fuck on
Sure. Like an artificial heart is artificial. Doesn’t mean it’s not really useful and amazing tech. Every tech we make is artificial if you think about it.
And what does “… are a chipper and solvent of competency…” mean?Doesn’t mean it’s not really useful and amazing tech.
you’re not tall enough for this ride
the fuck is bringing all you shitheads to this thread of all things?
I never claimed it wasn’t artificial, my argument is that it’s not intelligent.
And what does “… are a chipper and solvent of competency…” mean?
My observation from having to deal with entities which have had exposure to LLMs, is any competency or credibility they may have had, is eroded proportionally to their level of exposure. It’s equivalent to experiencing a traumatic brain injury.
Losing cognitive ability is bad enough, but it’s devastating when they never possessed it to begin with.
Its a dev tool.
Local models are fantastic homelab managers, way better than me
not suboptimal management of my home lab…anything but that…
the only dev tool here is you
Skill is acquired with use over time.
LLMs can be scalpels, but they’re currently being used as sledgehammers. Wielded in the right hands, they can appear to work magic, but they haven’t been available long enough for anyone to legitimately claim meaningful competence.
And none of it can possess or present comprehension, which for me is a requirement for intelligence.
LLMs can be scalpels
As in better suited for performing very fine tasks? It’s literally the opposite, being everything machines is their main feature, either in terms of churning out endless text at the vaguest provocation or by being a natural language intermediary between the user and the actual scalpel-like tools, that it then wields like a sledge hammer by forcing endless iterations until something sticks.
Wielded in the right hands, they can appear to work magic
Wielded in the right situation more like, which varies between models and can’t be a priori known.
The big problem with AI is that it removes the need (or perceived need) to perform the task that AI is replacing. This means that using AI will gain you no skill or knowledge of that task, and the skills and knowledge in the task is the only way to verify if the AI is giving the correct output.
In short, AI is actively removing from it’s users the one thing needed for it to be used effectively.
I can imagine performing better when you use AI to help learn the material, but not when you use it to do the assignment. Using it to find primary sources, or help you get your bearings on a topic you’ve been struggling with, doesn’t seem that detrimental.
It’s just a tool. If you use it responsibly, it can be helpful; if you use it irresponsibly, you can cause problems.
“IT’S JUST A TOOL” I shriek as I cut my own leg off with a chainsaw
Mate, literally the ancient Greeks wrote shit down about how tools shape behaviour.
In English we have “to the man with a hammer everything looks like a nail”. Nothing is neutral, you’re not that foolish you’re wishcasting.
yet another tool has entered and left the thread
