Hacker News new | ask | show | jobs
by dumbfounder 1266 days ago
ChatGPT is just a tool, like a calculator. Calculators are banned in certain circumstances and required in others. At first they were probably seen as cheating, but then the system realized they could be used to enhance what kids learn. Why take away a tool that could be used everywhere in life? That's not how progress should work. The system should evolve to embrace such tools, but the obvious problem is that it just won't happen quickly enough, and in the short term they will be used for illegitimate purposes, and so we just ban.
5 comments

Did you even read the article? You completely side stepped the argument they are making:

> While the tool may be able to provide quick and easy answers to questions, it does not build critical-thinking and problem-solving skills, which are essential for academic and lifelong success

I think we really need to get on integrating technology like this into a curriculum which does build critical thinking and problem solving skills, since these kids are already inundated with AI generated content and it seems certain to increase several fold over the next few years.
But the article is wrong...
Ok, this is why I think it's wrong: ChatGPT absolutely does teach us critical thinking and problem solving skills. How to best leverage ChatGPT is absolutely a problem that needs to be solved, and it varies wildly on the use case. Critical thinking absolutely does need to be applied to what it produces. All it can do without critical thinking and problem solving skills is create commoditized responses, the same thing it produces for everyone else. That has no value anymore. HOW you leverage is insanely important.
The problem is that calculators are correct 100% of the time, ChatGPT is not. What makes it more difficult is that it's often really hard to tell when ChatGPT is correct or not.
The problem is, people assume the tools they use are correct 100% of the time.

An early example: http://files.righto.com/calculator/sinclair_scientific_simul... (see Bugs and limitations).

More contemporary (I think): https://apcentral.collegeboard.org/courses/resources/example...

Granted though, ChatGPT is confidently and convincingly wrong more often. I expect that will improve, but in the meantime I think it's an opportunity to question and evaluate the results rather than ban outright. That's a sticking plaster that hopefully needn't stay on for long.

For whom is that a problem, in this case? If you have a hallucinating computer do your test and it doesn’t work out for you, that’s on you.
I think it boils down to the (intellectual) effort needed to synthesise your thoughts, which is a skill that leads to deeper understanding.

I'd say ChatGPT is a bit better as a tool than copying from a colleague, just because you still have to double check for correctness. However, does it rob students of that opportunity to think deeply and therefore learn? I'm not sure to what extent, but I'd imagine that there's an impact to _some_ extent.

ChatGPT is also going to get better, and will eventually be pretty accurate (or, at least as accurate as a student of a subject may be). My stance is that in school, you should be learning to think instead of memorising. If the goal is to memorise, however, then use ChatGPT isn't that harmful (you'll still fail for memorising the wrong information).

Overall I'd treat ChatGPT as a tool, and give students the facts: first gain a deep understanding of a subject so you can learn to verify, then use the tool later in life.

I'm also mindful of the fact that if an AI is not guaranteed to be correct, which is the case right now, then students now have an incentive to _understand_ the subject so they can verify, which might actually be a net good?

This same argument is valid for people. That does not mean people are useless, for example as assistent. Maybe we should just use that knowledge and still try get the best out of it.
Neither is Wikipedia. While some vandalism might be obvious, more nefarious edits might not be. It's still an extremely useful tool.
Not all tools are equal. Google Translate is a tool but it's not allowed in your language class, how surprising.
I see ChatGPT more akin to Wikipedia and Google than a calculator. We don't consider Wikipedia or Google as much of a tool as they are a knowledge reservoir of truthiness.
Have you used it? It can generate code. It can correct grammar. It can create Haikus (poorly). It can give you ideas. It is NOT just a knowledge base.
It's absolutely a knowledge base. The "training data" is its raw database which is then preprocessed to speed up the query engine. Then it does some additional processing to return amalgamated results from that database when presented with a query, which for some reason they call a "prompt." It's clever, but when you view the training data as part of the program source code, which it absolutely is, the information entropy of its output is miniscule.
Google can generate code.

- Google: How do I ... - visit stack overflow - copy/paste - code generated.

There's more steps, but how is this different? Chat GPT though will go the extra mile and actually EXPLAIN what each bit does usually. It's not always accurate but neither is stackoverflow.

It’s different in that the text that’s displayed in a chatgpt response never existed anywhere before you asked the question.

That’s what « generating » means, in that context.

What is a "knowledge reservoir of truthiness"?
The only knowledge this system knows is whatever it can scrape and regurgitate that appears truthful. A whole bunch of positively true, but also many cases of falsely positive.

This system can’t prove its own work (at least not yet) before publishing the results. It just publishes, which is similar to Wikipedia and Google.

The school administration.
The difference is that calculators have been around for decades and everyone is familiar with them, but chatgpt has been around a few weeks.