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

How University Students Use Claude

anthropic.com

Post reply on HN
Print view View original

561–570 of 601 posts

Re: How University Students Use Claude

#561

Post by dankwizard »

jplusequalt wrote: post #541
dankwizard wrote: post #326

Earlier quoted context omitted.

Your opinion is wild. Hmm, millions of humans are spending a bulk of their lives plugging away at numbers on a screen. We can replace this with an AI and free them up to do literally anything else. No, let's not do that. Keep being slow ineffective calculators and lay on your death bed feeling FULFILLED!

>We can replace this with an AI and free them up to do literally anything else. I would happily support automation to free myself, and others, from having to work full-time. But we live in a capitalist society, not StarTrek. Automation doesn't free up people from having to work, it only places people in financial crisis.

They'll get different jobs bud.

Re: How University Students Use Claude

#562

Post by guappa »

pc86 wrote: post #472
guappa wrote: post #466

Earlier quoted context omitted.

That's how it is in Italy. And that's why Italy is behind every other country in education. Because it hasn't yet made graduating as easy as it is in other places.

Well graduation rate is a pretty terrible way to grade education, especially country to country. You could have 100% graduation rate today by just passing everyone - that's basically what we have in primary education and there was an article here just last week about how most college students are functionally illiterate.

In sweden until high school it's literally impossible to fail. There's no grades and no way of failing anyone.

Then they suddenly become kinda stricter in high school, where your results decide if you can go to university and which.

But I've been to one of the top technical universities and compared to Italy it was very easy. It was obvious the goal was to try and have everyone pass. Still people managed to fail or drop out anyway, although not in the dramatic numbers I saw in Italy for math exams.

Re: How University Students Use Claude

#563

Post by sally_glance »

nixpulvis wrote: post #549
sally_glance wrote: post #480

Earlier quoted context omitted.

I agree that you should learn the fundamentals before taking shortcuts. I just don't view it as the universities' job to repeatedly remind their students of this, that's elementary/high school style. In universitiy, just give them hard problems requiring fundamental knowledge and cross-checking capabilities but don't restrict their tools.

I TA'd for the fundamentals of computer science I in college. In addition to being a great class for freshman, teaching it every year really did help keep me sharp. High schools are a long way off from that level of education. I took AP CS in highschool and it was a joke by comparison. Of course YMMV. The best highschool CS course might be better than the worst university level offerings. We would always have know it…

We could keep the same teaching offerings, my main gripe is with the assignments/examinations. It just feels wrong to complain about students using AI while at the same time continuing to hand out tasks that are trivial to solve using AI.

I also worked for the faculty for the better part of my university studies, and I know that ultimately changing the status quo is most likely impractical. There are not enough resources to continuously grade open-ended assignments for so many students and they probably need the pedagogical pressure to learn fundamentals. Still makes me a bit bitter from time to time.

Re: How University Students Use Claude

#564

Post by Ray20 »

jplusequalt wrote: post #542
Ray20 wrote: post #509

Earlier quoted context omitted.

>If we free up everyone to do anything Not anything, but something useful. And in exchange for that useful they'll get resources (which will become more abundant).

>something useful Where is the existing work these people would take up? If it doesn't exist yet, then how do you suppose people will support themselves in the meantime? What if the new work that is created pays less? Do you think people should just accept being made obsolete to take up lower paying jobs?

>Where is the existing work these people would take up? If it doesn't exist yet, then how do you suppose people will support themselves in the meantime?

Everywhere in human society. "Jobs" is literally when you do something that someone needs, so that in exchange they do something that you need. And in human society, because of AI, neither people’s needs, nor the ability to satisfy them, nor the possibility of exchanging them will suddenly disappear. So the jobs will be everywhere.

>Do you think people should just accept being made obsolete to take up lower paying jobs?

Let's start with the fact that on average all jobs will become higher paying because the amount of goods produced (and distributed) will increase. So the more correct answer to this question is "What choice will they have?".

AI will make the masses richer, so society will not abandon it. Subsidize their obsolete well-paid jobs to make society poorer? Why would anyone do that? So the people replaced by AI will go to work in other jobs. Sometimes higher paying, sometimes lower.

If we are talking about real solutions, the best alternative they will have is to form a cult like the Amish did (God bless America and capitalism), in which they can pretend that AI does not exist and live as before. The only question in this case is whether they will find willing participants, because for most, participation in such a cult will mean losing the increase in income provided by the introduction of AI.

Re: How University Students Use Claude

#565

Post by Ray20 »

hackyhacky wrote: post #514
Ray20 wrote: post #509

Earlier quoted context omitted.

>If we free up everyone to do anything Not anything, but something useful. And in exchange for that useful they'll get resources (which will become more abundant).

Huh? If AI can do any job cheaper and better than a person can, why would anyone hire a person? What "useful" skill could a person exchange for resources in an era when computers write code, drive cars, fight wars, and cook food?

But you answer your own question: the only situation in which it makes no sense to hire another person to satisfy a need is when that need has already been satisfied in another way.

And if all needs are already satisfied... Why worry about work? The purpose of work is to satisfy needs. If needs are satisfied, there is no need for work.

Re: How University Students Use Claude

#566

Post by davedx »

Zanfa wrote: post #422
enjo wrote: post #297

Earlier quoted context omitted.

> it's literally "let me paste the assignment into ChatGPT and see what it spits out, change a few words and submit that". My wife is an accounting professor. For many years her battle was with students using Chegg and the like. They would submit roughly correct answers but because she would rotate the underlying numbers they would always be wrong in a provably cheating way. This made up 5-8% of her students. Now she…

ChatGPT is laughably terrible at double entry accounting. A few weeks ago I was trying to use it to figure out a reasonable way to structure accounts for a project given the different business requirements I had. It kept disappearing money when giving examples. Pointing it out didn’t help either, it just apologized and went on to make the same mistake in a different way.

> me, just submitted my taxes for last year with a lot of help from ChatGPT: :eyes:

Re: How University Students Use Claude

#567

Post by dkersten »

hackyhacky wrote: post #280
dkersten wrote: post #265

Earlier quoted context omitted.

”I don't know how to grow crops, build a house, tend livestock, make clothes, weld metal, build a car, build a toaster, design a transistor, make an ASIC, or write an OS. I do know how to write a web site.” Sure. But somebody has to know these things. For many jobs, knowing these things isn’t beneficial, but for others it is. Sure, you might be able to get a job slinging AI code to produce CRUD apps or whatever. But…

> Sure. But somebody has to know these things. For many jobs, knowing these things isn’t beneficial, but for others it is. I think you're missing the point of my comment. I'm not saying that human knowledge is useless. I'm specifically arguing against the case that: > The irreducible answer to "why should I" is that it makes you ever-more-increasingly reliant on a teetering tower of fragile and interdependent supply…

You’re absolutely right. But my point still stands, too, which is that despite being irreversibly dependent on supply chains, doesn’t mean we are redundant. We still need people at all levels of the supply chain.

Maybe it’s fewer people, yes, but it’ll take quite a leap forward in AI ability to replace all the specialists we will continue to require, especially as the less-able AI makes messes that need to be cleaned up.

Re: How University Students Use Claude

#568

Post by never_inline »

sally_glance wrote: post #213
dtnewman wrote: post #14

> A common question is: “how much are students using AI to cheat?” That’s hard to answer, especially as we don’t know the specific educational context where each of Claude’s responses is being used. I built a popular product that helps teachers with this problem. Yes, it's "hard to answer", but let's be honest... it's a very very widespread problem. I've talked to hundreds of teachers about this and it's a ubiquitous…

I think this is a structural issue. Universities right now are trying to justify their existence - universities of the past used to be sites of innovation. Using ChatGPT doesn't dumb down your students. Not knowing how it works and where to use it does. Don't do silly textbook challenges for exams anymore - reestablish a culture of scientific innovation!

> Using ChatGPT doesn't dumb down your students. Not knowing how it works and where to use it does.

LLMs can't produce intellectual rigour. They get fine details wrong every time. So indeed using ChatGPT for doing your reasoning for you produces inferior results. By normalising non-rigorous yet correct sounding answers, we drive down the expectations.

To take a concrete example, if you tell a student to implement memcpy with chatgpt, and it will just give an answer which uses uint64 copying. The student has not thought from first principles (copy byte by byte? Improve performance? How to handle alignment?). This lack of insight in return to immediate gratification will bite later.

It's maybe not problem for non-STEM fields where this kind of rigor and insight is not required to excel. But in STEM fields, we write programs and prove theorems for insight. And that insight and the process of obtaining it is gone with AI.

Re: How University Students Use Claude

#569

Post by amanaplanacanal »

pc86 wrote: post #474
pja wrote: post #386

Earlier quoted context omitted.

The promise of the expansion of this kind of tutorial teaching to everyone via AI is great. The problem is keeping students honest with themselves.

At the end of the day you can't force people to learn if they don't want to. As a society we need to be okay with failing people who deserve to fail and not drag people across the finish line at the expense of diluting the degrees of everyone else who actually put in effort.

I'm not sure why we care about the degree. Employers care about the degree, but they aren't paying for my education.

The students who want to learn, will learn. For the students who just want the paper so they can apply for jobs, we ought to give them their diploma on the first day of class, so they can stop wasting everybody's time.

Re: How University Students Use Claude

#570

Post by andai »

idonotknowwhy wrote: post #554
jjmaestro wrote: post #443

Earlier quoted context omitted.

> They can do it (surprisingly well, once you disable the friendliness that prevents it) ... Interesting! :D Do you mind sharing the prompt(s) that you use to do that? Thanks!!

Have you tried Deepseek-R1? I run it locally and read the raw thought process, find it very useful (can be ruthless at times) seeing this before it tags on the friendliness. Then you can see it's planning process to tag on the warmth/friendliness "but the user seems proud of... so I need to acknowledge..." I don't think Gemini's "thoughts" are the raw CoT process, they're summarized / cleaned up by a small model befo…

That's fascinating. I've been trying to get other models to mimick Gemini 2.5 Pro's thought process, but even with examples, they don't do it very well. Which surprised me, because I think even the original (no RLHF) GPT-3 was pretty good at following formats like that! But maybe there's not enough training data in that format for it to "click".

It does seem similar in structure to Gemini 2.0's output format with the nested bullets though, so I have to assume they trained on synthetic examples.

Post reply on HN