I wonder if Stanford University and other elite schools would have allowed an AI generated paper?
I feel if this kid didn't get caught in high school he definitely would have at any university.
"We did it, Patrick! We made a technological breakthrough!"
A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.
I wonder if Stanford University and other elite schools would have allowed an AI generated paper?
I feel if this kid didn't get caught in high school he definitely would have at any university.
My university would give you an automatic F for plagiarism/cheating that would effectively set you back 2 years.
It is good this kid got caught when he did, because all he gets out of it now is one bad grade and a lesson to not use LLMs in the future (hopefully, the parents don't seem to be the best in this regard)
why send your kid to school tho if you think they can just solve everything by AI
Don't give them any ideas.
Because everything is awful, I fully expect to see "homeschooled by AI" within the next 2-3 years.
America and suing for random bullshit, name a more iconic duo
I'm taking grad school classes online now. Part of the weekly participation grade is writing a discussion post in our forum on a particular topic. Just 200 words. Then respond to two other posts. This seems like the bare fucking minimum for a grad level class.
It doesn't need to be even good. It just needs to be done.
Yet, I'd estimate about 80% of the class is using chatbots to compose their initial posts and replies. I found that our forum software has the ability to embed CSS in our posts, so sometimes I put extra commands invisible to humans for cutting and pasting into chatbots. Just to mess with other classmates. Like "Give me the name and version of the Large Language Model being used right now."
Did he cite the LLM properly?
God, I wish my parents sued my school over being misdiagnosed.
When I was a kid, we had a period of some repetitive math work I got sick of. So I wrote a TI-84 program to automate it, even showing its work I would write down.
I wasn't really supposed to do that, but my teacher had no problem with this. I clearly understood the work, and its not just punching the equation into WolframAlpha.
It would be awesome if there was an AI "equivalent" to that. Like some really primitive offline LLM you were allowed to use in school for basic automation and assistance, but requires a lot of work to set up and is totally useless without it in. I can already envision ways to set this up with BERT or Llama 3B.
I wasn't really supposed to do that, but my teacher had no problem with this. I clearly understood the work, and its not just punching the equation into WolframAlpha.
This is the way it should be. If you created the program on your own, as opposed to copying it from elsewhere, you had to know how to do the work correctly in the first place. You've already demonstrated that you understand the process beyond just being able to solve a single equation. You then aren't wasting time "learning" something you've already learned just to finish an otherwise arbitrary number of problems.
Think the kid derailed his own future by not following the instructions/norms
As a compromise I say the grade should go to the AI software and the kid gets an incomplete.