Noam Chomsky: The False Promise of ChatGPT

Re:Noam Chomsky: The False Promise of ChatGPT by teekert on Thursday March 09, 2023 @02:59PM (#35085844)

I was going to say the same thing, if you've interacted with it, in some depth, you know how human it may seem in one sentence then in the next completely an utterly proves itself to be a machine. Yet some people (some examples are well know) really project a human like mind onto the thing (as posted here before, this is also insightful [0]).

[0] https://nymag.com/intelligencer/article/ai-artificial-intell...

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by jamiek88 on Thursday March 09, 2023 @03:08PM (#35085978)

There are people who literally had pet rocks.

Humans can project feelings onto cars never mind something that can communicate with us!

Just look at Replika.

I'm not surprised people are projecting sentience onto these things. I am worried about the fall out though.

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by jiocrag on Thursday March 09, 2023 @03:12PM (#35086024)

Aren't we inherently projecting feelings onto anything that isn't inside our own direct experience? There is no way to confirm any alleged sentience outside of your own "feelings" is not an automaton, including other humans.

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by esailija on Thursday March 09, 2023 @03:22PM (#35086146)

It depends what you mean by confirm.

You can raise the bar for confirmation so high that you cannot confirm that you even yourself are not an automation. And you can lower the bar so that other humans are considered sentient.

But in the end, it is only about arbitrary decision where to place the bar for confirmation.

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by jiocrag on Thursday March 09, 2023 @03:59PM (#35086618)

How could we possibly ever test for "not feeling like an automaton?" The only possible test, it seems, is internal subjective experience. Even if something reported this, how would you ever verify it?

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by jamiek88 on Thursday March 09, 2023 @03:19PM (#35086100)

Well ultimately, yes.

Our brains are trapped inside a box of bone and only know what we simulate from our vision that is itself mostly 'upscaled' and milliseconds behind 'real time' as well as our other senses.

We take a LOT on faith and our systems are so flawed that magicians exist !!

It's amazing we can do anything.

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by hot_gril on Thursday March 09, 2023 @03:21PM (#35086140)

Yep, I'm convinced that cars are designed to have human-like faces, even if not consciously.

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by teekert on Thursday March 09, 2023 @03:31PM (#35086264)

It almost has to be, that's why small shopping cars (Twingo?) look "happy", cool BMWs look angry, aggressive.

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by Silverback_VII on Thursday March 09, 2023 @03:43PM (#35086397)

I would rather say animal-like but I'm sure these are conscious choices.

[ Parent | on HN ]

Re:Noam Chomsky: The False Promise of ChatGPT by hot_gril on Friday March 10, 2023 @05:46PM (#35102327)

At least with the 2000 Jaguar XKR (cat) or the 2000 Camaro (catfish).

[ Parent | on HN ]