Weekly Wikipedia Find: Chinese Room
Thought Problem! Thought Problem! Thought Problem!
The Chinese room argument is a thought experiment (or the German Gedankenexperiment) . On first read through, its full philosophical impact failed to hit me (my fault, not its). Still, it's no Turing test. That said, like the Turing test, it is concerned with artificial intelligence.
I'll attempt a quick run through of the actual problem-- without plagiarizing the Wikipedia entry (or the problem's originator, John Searle):
Imagine a computer. It's 1980, so imagine a 1980 computer. It shouldn't be too hard since you're a person in 1980. Imagine the year is 1980. Now a computer-- hopefully you know what a computer is. Yeah, those big room-sized machines with the punch cards and the more hassle then it's worth. Now, this computer, its input is Chinese characters and its output is other Chinese characters. To get this other from the non-other, it runs a program. This program works astonishingly well, so well as to pass the Turing test. That is, it convinces a Chinese speaker (human, naturally) that it too is human. There is the suggestion that the computer is an artificial intelligence and understands Chinese, understands like a human.
Now imagine, that instead of a program, the computer is with someone in a room, an English speaker (the computers are networked-- yes, like the internet. Oh, you don't know the internet? Geez, check out the rube from 1980.). Now this English speaker receives the Chinese speaker's input, consults a book on how to process the input to create the output, and does so. In this way, the English speaker manipulates the computer to pass the Turing test for the Chinese speaker. Without an ounce of artificial intelligence.
Did I mention that it is a criticism of the Turing test?
And that is it. The rest of the Wikipedia page is kind of extra-textual. It contains some background, some definitions, and most intriguing a lot of responses to the argument.
Week One: Lolita fashion
The Chinese room argument is a thought experiment (or the German Gedankenexperiment) . On first read through, its full philosophical impact failed to hit me (my fault, not its). Still, it's no Turing test. That said, like the Turing test, it is concerned with artificial intelligence.
I'll attempt a quick run through of the actual problem-- without plagiarizing the Wikipedia entry (or the problem's originator, John Searle):
Imagine a computer. It's 1980, so imagine a 1980 computer. It shouldn't be too hard since you're a person in 1980. Imagine the year is 1980. Now a computer-- hopefully you know what a computer is. Yeah, those big room-sized machines with the punch cards and the more hassle then it's worth. Now, this computer, its input is Chinese characters and its output is other Chinese characters. To get this other from the non-other, it runs a program. This program works astonishingly well, so well as to pass the Turing test. That is, it convinces a Chinese speaker (human, naturally) that it too is human. There is the suggestion that the computer is an artificial intelligence and understands Chinese, understands like a human.
Now imagine, that instead of a program, the computer is with someone in a room, an English speaker (the computers are networked-- yes, like the internet. Oh, you don't know the internet? Geez, check out the rube from 1980.). Now this English speaker receives the Chinese speaker's input, consults a book on how to process the input to create the output, and does so. In this way, the English speaker manipulates the computer to pass the Turing test for the Chinese speaker. Without an ounce of artificial intelligence.
Did I mention that it is a criticism of the Turing test?
And that is it. The rest of the Wikipedia page is kind of extra-textual. It contains some background, some definitions, and most intriguing a lot of responses to the argument.
Wikipedia by Week
Week Two: Ambrose BurnsideWeek One: Lolita fashion
Labels: artificial intelligence, chinese room, john searle, plagiarism, rubes, thought experiment, turing test, wikipedia
posted by justin | 3:00 p.m.
0 Comments:
Post a Comment
<< Home