Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

My little sister's use of ChatGPT for homework is heartbreaking

old.reddit.com

Post reply on HN
Print view View original

431–440 of 523 posts

Re: My little sister's use of ChatGPT for homework is heartbreaking

#431

Post by fivelorn »

AndroTux wrote: post #216
pembrook wrote: post #201

Earlier quoted context omitted.

Every study done on the topic of “people judging the stupidity of other people” always proves this comment incorrect. With or without AI, we have always thought we are of wildly above average intelligence and that “all the other people are brain dead zombies.” This is just a classic human bias since we don’t ever experience what goes on inside anyones head but ours. Everything is going to be just fine.

I don't even know how to respond to that.

You can't believe your lying eyes!

Re: My little sister's use of ChatGPT for homework is heartbreaking

#432

Post by fivelorn »

dfxm12 wrote: post #262

Counterpoint: maybe, if anyone is even cooked, it is just this one student and not an entire generation. Go back generations and it was always something else ruining the youth: not playing outside enough, Internet, videogames, hip hop, playing outside too much, TV, rock & roll, books, etc. In my own studies and brief time as a tutor, you get students like this who just see school as an obstacle to something else. Eve…

What happens if the entire generation is students like this?

Re: My little sister's use of ChatGPT for homework is heartbreaking

#433

Post by 3good5me »

dmos62 wrote: post #243

The kid doesn't care at all about the problems she's asked to solve. I considered that normal until recently. Now I'm increasingly surprised that no one bats an eye that we're forcing kids to learn. Naturally that has little success. I heard someone say "adapt the lesson to the student's excitement, not the excitement to the lesson". I think that's a foundation for something wholesome, though, admittedly, I struggle…

At some point, you need to learn how to do things that are hard, or unpleasant, or just downright boring. Not everything can be a MrBeast video, that just isn't how life works. We're raising kids who can't just sit with their own thoughts; they need constant stimulation. But guess what? Long division isn't stimulating. Taxes aren't stimulating. Cleaning the kitchen isn't stimulating. But you need to do those things anyway.

I find it hard to believe that it's mere coincidence that everything fell apart right after we invented these perfect little surveillance Skinner boxes, and then put overpriced Markov chains in them and unleashed them on children.

Re: My little sister's use of ChatGPT for homework is heartbreaking

#434

Post by sinuhe69 »

eYrKEC2 wrote: post #427

Why does this 6th grader have math homework? Finns have less time in school every day and they still do better than Americans.

I believe the PISA comparison in this context is flawed. To compare any statistics, we must control for other variables, otherwise any conclusion about causal effects and relationships could be deeply wrong. In the education context, the most important variables are the economic background of the households, the active involvement of the parents in the education of the children, the amount of time spending on the subjects (both in class and at home), the curriculum and methods of teaching.

Finland has only 5 millions population and a relatively high median income, the teacher:students ratio is much higher and the percentage of intact families are higher. Therefore, if we want to compare, we have to select better districts in the US, where the teacher:students ratio is similar high then see what the PISA scores will say. Only then, and looking deeper into the curriculum and teaching method, we can say whether homework is beneficent or not.

Re: My little sister's use of ChatGPT for homework is heartbreaking

#435

Post by hb53821 »

Peacefulz wrote: post #3

I'm not trying to justify, or infer here, but I read recently that around 20% of American adults are functionally illiterate. https://nces.ed.gov/pubs2019/2019179/index.asp I wonder how LLMs are going to affect these metrics.

That is terrifying. But what is the trend? Maybe it’s historically low. I tried Googling around and could only find recent data of that same 2019 report.

Re: My little sister's use of ChatGPT for homework is heartbreaking

#436

Post by TZubiri »

sophiebits wrote: post #418
TZubiri wrote: post #399

Earlier quoted context omitted.

"The shit version of chatgpt is free" Fixed that for you. And it will get shittier as investment dries up and the pressure to monetize increases.

4o-mini is more than capable of doing grade school homework.

It's the other way around, no model is capable (or should be put in charge of) doing grade school homework. So the pro version is less shit at it.

The fact that we use the LLM should always be preceded with, "I know I shouldn't be doing this but..."

Re: My little sister's use of ChatGPT for homework is heartbreaking

#437

Post by noufalibrahim »

The larger trend of "making things more user friendly and easy" is part of this problem.

I cut my teeth on early Linux with really poor driver support and had to figure out so many things just to get X to start or to get audio to play. This made me intimately familiar with the system and I had the confidence to explore things myself.If were to start off with a modern Ubuntu, I wouldn't be aware of any of this. Wouldn't know where the logs are, how to look at them, what a certain error meant etc. "A system error has occured. Click here to submit a bug report to Canonical" is just not quite the same as digging through some obscure log file and trying to create a mental model of how the distribution works before the next attempt at fixing the problem.

Things which require tinkering to get working are invaluable during the learning journey. Problems which require effort to solve are one example of this and a powerful solution that eliminates the effort and directly provides the reward is surely going to cripple learners.

In this specific case, I think the parents have "outsourced" education to the school/teachers/tools which is why this has happened.

Re: My little sister's use of ChatGPT for homework is heartbreaking

#438

Post by tasuki »

fivelorn wrote: post #424
Peacefulz wrote: post #3

I'm not trying to justify, or infer here, but I read recently that around 20% of American adults are functionally illiterate. https://nces.ed.gov/pubs2019/2019179/index.asp I wonder how LLMs are going to affect these metrics.

You need to be able to read to use an LLM (the second "L" is particularly salient).

Actually voice

Re: My little sister's use of ChatGPT for homework is heartbreaking

#439

Post by mmooss »

fivelorn wrote: post #428
mmooss wrote: post #211

Earlier quoted context omitted.

They haven't accepted Covid vaccinations, immigration, the election of Joe Biden, ... . The world has a long history of people determining their fate and not accepting things. It's just one political group that has embraced quitting and despair - they are laughable.

Who could possibly be miserable amidst disease and authoritarianism?

The question reveals the problem, a victim outlook. You have no responsibility or agency, and your feelings justify it. It's not only bs, it's unhealthy and self-defeating. The outcome is predictable.

Re: My little sister's use of ChatGPT for homework is heartbreaking

#440

Post by Terr_ »

TZubiri wrote: post #405
Terr_ wrote: post #350

Earlier quoted context omitted.

That analogy rests on some precarious assumptions, like: 1. That the "AI" being bandied about (read: LLMs) really is able to "kill" the other blocs. 2. That such an action will be safe from retaliation. 3. That software engineers actually control/own enough of everything to pull that "trigger" unilaterally.

> 3. That software engineers actually control/own enough of everything to pull that "trigger" unilaterally. Either you misunderstand the quote, or you haven't read the book or watched the series. George Martin's central motif of what power really is doesn't have a definite answer, several characters propose different theories, in this case Varys poses more of a question than an answer. You might say that the AI holds…

I slogged through the ~4,000+ pages of the 5 books--not that I consider myself a fan at this late date--and the subtext of my post was: "I don't think that thought-experiment maps usefully to anything around us right now, it barely did for the characters in a medieval fantasy."

All models are wrong, some are useful, I don't think this one is that useful, unless some researcher has a Technological Singularity in a Bottle somewhere.

Post reply on HN