It's far from proven or obvious whether involving an LLM in your thought process degrades your thought process.

theappsecguy 8 hours ago | parent | next [-]

It seems plenty obvious, but there's also scientific backing slowly catching up: https://www.media.mit.edu/publications/your-brain-on-chatgpt...

▲

fc417fc802 7 hours ago | parent | next [-]

It's not at all obvious because there's more than one way to go about it. Obviously entirely outsourcing is bad. Whereas working cooperatively seems highly beneficial to me.

Google search has been getting progressively worse for technical topics for at least the past decade. Now suddenly they started providing a free tutor capable of custom tailoring graduate level explanations of technical topics for me on demand. The difference is night and day.

▲

multjoy 7 hours ago | parent | next [-]

How do you know that the explanations are free from error?

▲

charcircuit 7 hours ago | parent [-]

You can still learn from sources that have errors. Many textbooks have mistakes and false information in them, but that didn't stop them from providing educational value to people.

▲

multjoy 7 hours ago | parent [-]

We're talking about LLM's that are designed to be confidently incorrect. Accuracy is a side-effect.

	▲	fc417fc802 7 hours ago \| parent [-]
		When textbooks are incorrect it is also with great confidence. If you can't spot logical inconsistencies in the material were you actually learning or merely memorizing?

▲

kelnos 7 hours ago | parent | prev [-]

Sure there's more than one way to go about it, but what matters is how people typically do go about it.

And certainly individuals can make their own decision to engage with an LLM in positive, self-thought-provoking ways, but it's still useful to understand how people generally do use them in the real world.

▲

wilg 7 hours ago | parent | prev [-]

That's about essay writing exclusively.

▲

kelnos 7 hours ago | parent | prev | next [-]

Sure, so we shouldn't assert that with confidence, but I think it's safe to guess that, for most people's use, that is probably the case.

Yes, some people (see some sibling commenters) do engage with an LLM in ways that might make them more thoughtful, but I have a hard time believing that's the common case.

▲

justinnk 7 hours ago | parent | prev | next [-]

I think it really depends on the how. Engaging with it in a socratic debate-style argument [1] if no fellow human is available might very much support your thought process. On the other hand, just obtaining the solution to one‘s homework/problem/task/… won‘t be very beneficial for one’s development. The latter is sadly much more convenient and probably accounts for most of the usage. I remember a saying about the mind being a muscle: in order to keep it in good shape, you have to use it actively.

[1] https://en.wikipedia.org/wiki/Socratic_method

▲

kl33 6 hours ago | parent | prev | next [-]

Long-time lurker.

Personally I stopped using LLMs much from around 6 months ago. I was using them regularly prior to that.

I noticed these dimensions of myself increased:

- Patience - Focus - Ability to hold concepts and reason for longer

and other related qualities improved.

My personal experience tells me they do degrade or hinder oneself from operating maximally. Some may be more sensitive than others - we aren't all the same.

But one thing for sure - younger generations will be more sensitive as they are already exposed to products that are designed to erode their self-control.

▲

AirGapWorksAI 8 hours ago | parent | prev | next [-]

Agreed. In my case, I think I have found the opposite. At least, I find myself thinking hard about things more, now that I have started working hand in hand with AIs on different projects. Which is probably enhancing my cognitive ability, not degrading it.

▲

andy99 8 hours ago | parent [-]

This captures the problem, the sycophancy / preference optimization deludes people into thinking they’re on to something and posting things that don’t contribute to the discussion. It’s the “I drive better when I’m drunk” syndrome, it’s better just to outright ban it than to leave it to people’s judgement.

	▲	wilg 7 hours ago \| parent [-]
		The point is we don't know whether that's true, only that some people think it's true, which is not interesting.

▲

goatlover 7 hours ago | parent | prev [-]

It degrades my thought process reading it when I'm expecting human comments. If I want to converse with an LLM, I can do that already.