Sounds like a lot of humans I know
Descartes in 1637, on how language models would be able to generate realistic-sounding text, but would not be able to show understanding under probingpic.twitter.com/P8degZLj01
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Pre Turing.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Sounds as Searle’s Chinese Room thought experiment. However the date is 1637..
-
The problem with the Chinese room is that the criticism breaks down when we endow a machine with the same level of contextual information that a person has. In that sense either we are all Chinese rooms or a Chinese room isnt problematic after all.
- Show replies
New conversation -
-
-
He definitely would think that considering that he equated animals with machines. In the Descartes view all animals were reflex driven machines incapable of feeling pain.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Do you think that Descartes's language model here is based on the concept of a private language? An alternative was proposed by Wittgenstein where he argues that language is essentially social. Ref: https://plato.stanford.edu/entries/private-language/ …https://en.wikipedia.org/wiki/Private_language_argument …
-
How would the meaning be derived for a generated text, hypothetically (realistic, or otherwise)? Is it from an a-priori thought, based on intent and a private language of the generator of the text, or is the meaning conveyed after the text is socially surveyed?
- Show replies
New conversation -
-
-
He was right so far. Brain computer interfaces may change that, or at least make it almost unnoticeable...
-
We already have machines that can do this though... Us Humans
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.