You are not logged in.

#26 2024-05-15 14:36:01

ewaller
Administrator
From: Pasadena, CA
Registered: 2009-07-13
Posts: 19,904

Re: Gentoo decided to not allow AI generated code contributions

Trilby wrote:

Nephilim lizard people aliens who have infiltrated all our governments.

But it would answer a lot of questions.  Adventures of Buckaroo Banzai Across the 8th Dimension.


Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
Sometimes it is the people no one can imagine anything of who do the things no one can imagine. -- Alan Turing
---
How to Ask Questions the Smart Way

Offline

#27 2024-05-15 14:39:27

Trilby
Inspector Parrot
Registered: 2011-11-29
Posts: 29,744
Website

Re: Gentoo decided to not allow AI generated code contributions

Exactly.  They're coherent theories because they are internally consistent and could answer a lot of questions.  But that pesky bit about evidence to whether it's actually true ... that's why - while deliberately colorful - my comparison is not hyperbole.  I see speculating about the functioning of brains to be just as silly as speculating about these alien overlords.

Last edited by Trilby (2024-05-15 14:40:02)


"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman

Offline

#28 2024-05-29 21:14:06

libXq
Member
Registered: 2024-05-08
Posts: 15

Re: Gentoo decided to not allow AI generated code contributions

Trilby wrote:

... I challenge you to present even a basic concept of how humans "think" ...

I am aware of the concepts you mentioned. Even though I am not an expert in this field, you raised some valid points that I find myself thinking about quite often. In my personal opinion, intelligence seems to be a very controversial topic. This starts at the level of human perception, and the same goes for the definition of life. For example, the definition of life "what it is and whether it has any purpose" can be difficult to measure, as can intelligence. Science even differentiate between intelligence and consciousness, which is a topic adding another layer of complexity to the discussion.

There are some definitions of life available, but even if accepted by some biologists, there are still no universally accepted definitions, as is the case with intelligence. But do we need a definition at all? Don't we all already have our own sets of definitions for everything (shaped by language, individual preferences, genetics, and societal background) that differ so much that, in the end, it is always about opinion, whether in science, academia, or society in general?

But I can turn the question back to you: Do you think that your way of thinking is just a sequence of measurable inputs (such as your senses eg seeing, smelling, ..) and predictable outputs? Even if you could sense everything at a given moment (access to data), and you deliberately have well-defined outcomes for everything you sense (training on all possible outputs and choosing preferences based on the rules of the system, what might be considered good or bad), is that the concept of intelligence? Just imagine you don't have the ability to dream "not in the sense of sleeping", but dreaming of something you've never smelled, touched, or tested.

This analogy can be applied to large language models. These models process measurable inputs and produce outputs based on patterns they have been trained on.

Last edited by libXq (2024-05-29 21:27:38)

Offline

#29 2024-05-29 21:55:50

Trilby
Inspector Parrot
Registered: 2011-11-29
Posts: 29,744
Website

Re: Gentoo decided to not allow AI generated code contributions

libXq wrote:

But I can turn the question back to you: Do you think that your way of thinking is ...

What I personally and individually think about any such questions is irrelevant.  That's my point.  We all have our intuitive senses about all of these questions just as we do about AI.  But our intuition is only good at questions we face and have faced on a regular basis throughout our evolutionary history (and even then it is quite prone to systematic errors).  Our intuition is less than useless when it comes to quantum physics, the functioning of biological brains, and to whether or not AI thinks.  We cannot rely on our intuition for such topics (and probably many others) - instead we need to rely on logic and evidence.

You made a positive assertion of a difference between AI thinking and human thinking.  If that is just your intuition, so be it - but then it is worth less than nothing.  I was asking for logic or evidence to back up your claim.  Turning the question around to ask about my intuition is silly: my intuition on such topics is also worth less than nothing.

As a neuroscientist I am familiar with quite a lot of evidence about human and animal brains.  And with that evidence I can reject some poorly formed hypotheses - but the existing evidence is not sufficient (yet) to have any reliable grand overall picture of how humans "think".  So to make positive assertions about similarities or differences between human, animal, and computer "thinking" is just silly.  It's wild speculation void of fact, evidence, or logic.  I will not partake.

I reject your assertion as it lacks any foundation.  I need not be able to provide any alternative assertions to do so.

I will - for illustrative purposes - acknowledge that my intuition is inclined to say human thinking and AI thinking is quite different.  But as a scientist I question my intuition.  As someone interested in history - particularly history of science - I am also well aware that for much of human history every sane person's intuition told - and they were quite confident in believing - that people of different races or ancestries thought different and / or that some were incapable of rational thinking.  This is an abhorrent pattern of human history; we should learn from it.

Last edited by Trilby (2024-05-29 22:02:26)


"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman

Offline

#30 2024-05-29 23:04:32

libXq
Member
Registered: 2024-05-08
Posts: 15

Re: Gentoo decided to not allow AI generated code contributions

As you consider yourself a Scientist, you should also question the concept of evidence. As you already mentioned in history, most geniuses were initially rejected by evidence until the evidence suddenly proved false, and the previously rejected ideas became the leading mindset. We certainly have different backgrounds, but from my perspective, from the IT side, these technologies do not seem to follow what I would call intelligence.

Offline

#31 2024-05-29 23:34:50

libXq
Member
Registered: 2024-05-08
Posts: 15

Re: Gentoo decided to not allow AI generated code contributions

But to be honest, I think you're going way too far. You can't just dismiss someone's argument as silly. You reject discussion, and then you talk about racism connected to rejection. I think working as some sort of scientist seems to be quite easy.

Offline

#32 2024-05-29 23:44:06

Trilby
Inspector Parrot
Registered: 2011-11-29
Posts: 29,744
Website

Re: Gentoo decided to not allow AI generated code contributions

libXq wrote:

I think working as some sort of scientist seems to be quite easy.

I think your mom is easy.  And because of that, the world gets you.


"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman

Offline

#33 2024-05-30 00:56:59

ewaller
Administrator
From: Pasadena, CA
Registered: 2009-07-13
Posts: 19,904

Re: Gentoo decided to not allow AI generated code contributions

And this was just getting interesting.  I follow the argument, but the last three posts set off the 'report' alarms.  Not sure I totally agree, but I have come to accept the increasing sensitivity of things.  And, it is starting to get personal from both sides.  How can it not when we are talking about ethereal perceptions about how we think? 


It is clear this is no longer about Arch Linux, so I declare that the train left the rails and it is time to bring this thread to a close.

Last edited by ewaller (2024-05-30 00:58:28)


Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
Sometimes it is the people no one can imagine anything of who do the things no one can imagine. -- Alan Turing
---
How to Ask Questions the Smart Way

Offline

Board footer

Powered by FluxBB