Being able to distinguish between useful, valid information and whatever YouTube video the search engine happened to turned up is going to be an even more important skill for those kids in school now than it was for those of us who were in college when Google was a hot new startup.
For those of you who didn't read the whole post (and it was long, so I kind of understand), Mr. Devereaux made an aside about his belief in the continuing value of initially learning how to do arithmetic without a calculator despite their easy availability over the past few decades, an opinion I've always shared.
Before reading this post, I still believed the same about the value of learning how to write essays ("delivery boxes for thoughts" was his expression, I think) and will make sure that my kid can write one with just a pencil and paper, even though he'll also be able to use whatever technical assistance is available in 10-15 years. He's learning to draw and make letters with crayons and pens before I'll let him spend a lot of time with my iPad; he's sussed out how that worked just by watching me, so I'm not concerned about a technology gap with his future classmates.
This post gives me something to forward to my non-technical but curious friends when they ask about ChatGPT and similar.
not that i disagree, but what would you have it change to?
The fact that there's no need to memorize anything any more, means that any tests which allows the use of tech (like the internet, or chatGPT) to retrieve data, means that students no longer need to commit things to long term memory.
However, having the capability, and be able to recall facts when needed, and at high speed, is something that is foundational to higher creative thoughts.
This higher, creative thought, is not really easy to test, so the memorization is the proxy.
I do not know what form education (and the testing of it) would take on, if tools like chatGPT is allowed to be used.
For example, I had to memorize the structural formulas for all amino acids in university. This does seem a bit useless at first, but is actually very important the moment you work with protein sequences or structures. You might not need the exact structure, but if you read about a specific important residue in a protein or a mutation in one you need to understand the properties of the involved amino acids to make sense to this. And if you had to look that up every time you'd never get through a paper.
Teachers aren’t interested in a recount of the battle of so-and-so but in training you to gather knowledge, structure your thoughts and express them clearly.
You can only learn that by doing. A chatbot bypasses the learning process, so you will have neither gained subject knowledge nor methodical one.
https://www.nytimes.com/1991/09/29/opinion/the-calculator-cr...
I'm pretty sure that a "sufficiently smart" chatbot (or maybe even an extra dumb one) is a useful tool "in training you to gather knowledge, structure your thoughts, and express them clearly". I've found it remarkably useful for clarifying my thoughts, considering alternative arguments, and general tomfoolery that can spark creativity.
I am mostly worried about young people who will grow up relying too much on ChatGPT, what will they do when they do not have a bot hand-holding them through some complicated idea? And if this kind of bots become so ubiquitous, what is the place for humans?
When calculators became wide-spread, we calculated a lot more. When LLM become wide-spread, we will.. Think more? I seriously don't know.
I certainly benefit greatly already from using LLMs to accomplish a number of tasks. I think the answer on where the responsibility lies depends greatly on your view of the same sorts of questions around auteur theory - is the director responsible for the quality of the film? Or is it the writer of the screenplay? What about the cast, or the producers? Is Microsoft the author if you write a novel in Word, without scribing the lines onto the page yourself? I think it's going to be very interesting to see how all of this plays out, and where the lines are drawn. I suspect that what is causing concern now will, in ten years perhaps, be normal, obvious and not even discussed.
I agree, and that is the issue. People like us can use LLMs effectively because we are already capable of expressing our thoughts in a decent manner and we can recognize when the output does not make sense. But to know whether the results can be trusted or not, you already need to be one level above that. If one is not capable of producing a coherent argument on their own, how can they evaluate whether an argument they hear is itself coherent? And if one, say because of lazyness, relies on LLMs from their childhood to fill in all the difficult steps, how will they learn how to do it on their own? Practising has always been the best way to learn things.
> I think it's going to be very interesting to see how all of this plays out, and where the lines are drawn. I suspect that what is causing concern now will, in ten years perhaps, be normal, obvious and not even discussed.
Well said.
But this is somewhat off topic because the article is talking about essays in the context of a university, not school, education.
(Not to mention that parents are still at least partially responsible for their children's education.)