Anonymous #2 4w

It still suggests a lack of confidence in your own work. Mistakes are human, let yourself be human.

35

Anonymous #1 4w

Don’t use it. Period. End of discussion. You don’t need it. There are plenty of other resources out there to help fact check or look things up.

53

Anonymous #3 4w

This is flawed logic. To know if you got something correct you should check to see if you get it right. Not checking your work is worse than using an external tool to assist in this. Yes mistakes are human but you are never gonna learn if you do nothing to correct said mistakes.

13

Anonymous OP 4w

What if I am an international student and I use it to correct grammar, translate and help me w like structure and make texts seem more like they were written by a native speaker

0

Anonymous #1 4w

I know is still bad but I feel like normal translators don’t capture my ideas well or uses vocabulary that I don’t like

1

Anonymous #5 4w

Checking your own work or asking a peer/mentor to look it over has existed for thousands of years my dude. If you’re using AI for it you’re just lazy and uncaring about the horrible impact AI is already having on humanity and our planet, especially since we’re more interconnected than ever and could literally just go ask Reddit about it.

1

Anonymous #1 4w

you might benefit from trying therapy. good luck buddy you got this

20

Anonymous #1 4w

Grammarly perhaps? There’s still AI in it but it’s like an editing tool and I feel it’s not as bad as asking chat for everything. And if you lock in on the suggestions it’s offering you can learn some grammar along. That’s the free version. Paid it can help you with voice or you’ll j have to work a little harder with a dictionary or asking a friend to help make it sound like you want

12

Anonymous #2 4w

Depending on what work if it’s like math problems or sm with a definite answer sure but if it’s an essay or written questions I feel like your less likely to learn from your mistakes or can do a peer review. And like 3 said god forbid it’s not fully correct and you wait for feedback

-2

Anonymous #10 4w

LLMs are horrible at math! It's a language model, not an actual calculator.

8

Anonymous #1 4w

For this, I’d say you’re probably okay. The issue comes when people use it to replace critical thinking, which doesn’t seem like what you’re doing here. It probably doesn’t help your language learning journey, but if, for the time being, daily functionality is more important than working towards mastery, I’d say you’re totally fine. Just don’t start asking it to plan your day for you or something.

5

Anonymous #2 4w

If you’re a college student there are usually resources for that. I know my school has a writing center where you can get someone to read over your paper and a peer study group for STEM stuff and labs. Having a human to talk it through with will also help you learn more. It’ll also be more reliably correct than ai, not to mention better for the environment.

1

Anonymous #13 4w

That final session was probably the most important one since much of their conclusions are based on it. It had 2 conditions, so N = 9 for each: People who had been using LLM for the previous 3 sessions switched to brain-only to write essays (on topics they already wrote about), and people who had been using brain-only switching to LLM to write an essay in the same way. The brain to LLM group had the highest neural connectivity based on EEG. That is essentially the conclusion

-3

Anonymous #13 4w

The constraints for the previous 3 sessions in the LLM only group were harsh too. They had a 20 minute time limit. They were time-pressured to just copy and paste the output of the LLM instead of engaging with it with meaningful constructive questions. Of course they will struggle to remember everything they copied and pasted and will have less ownership over their essay. Of course they will use less neural connections to do so

-3

Anonymous #13 4w

Everyone had that time limit but it still pressures the LLM group to just copy and paste. The authors themselves also note that there were differences in cognitive load based on how users approached the LLM (even despite the time limit) :

-2

Anonymous #13 4w

They also loosely defined “Cognitive engagement” and based it largely on just EEG signals. And yeah the LLM group had the lowest during the 3 writing sessions, while the brain-only group had the highest. A group that was allowed to use a search engine was in between

0

Anonymous #13 4w

This study really just shows that you will use less neural connections (only in ways reflected on an EEG) if you use LLMs in a specific way that was encouraged by the constraints of the experiment conditions. Like using a search engine to copy and paste something from wikipedia uses less neural connectivity than sitting with google docs open and not using the internet at all. And even then the sample size is tiny

-1

Anonymous #13 4w

No cognitive ability of any kind was assessed, only eeg signals. There were no subtasks within the essay writing process to assess dimensions of cognition. This study can’t be used to make huge claims like LLMs will atrophy your brain or neural connectivity. At best you can infer yourself that using it in a copy and paste way will make you think less and get worse at understanding how to approach problems, but that’s exactly why we were taught not to copy and paste from wikipedia

-1

Anonymous #13 4w

I personally think that using AI in a copy and paste way instead of in a critical way probably does erode your critical thinking skills. Doing it nonstop probably does make you overall dumber. But that was always true. Thats like saying that reading books often is good for your attention span and critical thinking. We know. But people saying it causes brain damage or unequivocally makes you overall dumber are just as brainrotted because they don’t have the attention span to read past headlines

0

Anonymous #14 4w

And you’re fully rotted if you think AI isn’t damaging the brains of those who use it.

7

Anonymous #13 4w

Absolutely insane that you people will upvote misinformation and misrepresented science, and downvote the actual reality of the study. Its fine to have your own agenda but use your brains and find evidence that isn’t taken out of context or misrepresented

6

Anonymous OP 4w

Completely ignoring all nuance and what the science actually says just because you don’t like AI in a general sense is brainrot

0

Anonymous #12 4w

Saying "my school" on the school yak is such a funny thing to me, no hate, just thought it was funny

4

Anonymous #13 4w

There are SO many reasons to hate AI. The fact that you don’t shows how far gone you are.

1

Anonymous OP 4w

Then present those reasons genuinely and don’t lie and spread misinformation. Thats all im saying. And i never said i don’t hate it or hate aspects of it or how its used commercially for slop. I just think its a complex topic and needs to be considered as such. Lying about it just cheapens your argument

1