3 ms·
What does the Turing test really mean? (2009) [pdf]
- jjtheblunt 4y agoI think the Turing test is more interesting than what people normally believe. I think Turing's test involved the ability of the machine to deceive the human, as well, regarding some aspect (perhaps gender?) of the machine's faux human persona. This wikipedia article touches on the subject of co-opting the phrase "Turing Test". https://en.wikipedia.org/wiki/Turing_test#Imitation_game_vs https://en.wikipedia.org/wiki/Turing_test#Imitation_game_vs.... And the book below is really interesting. https://en.wikipedia.org/wiki/Alan_Turing:_The_Enigma https://en.wikipedia.org/wiki/Alan_Turing:_The_Enigma
- motohagiography 4y agoI used to say the Turing Test wasn't a philosophical problem, but a measure of an indifference curve from economics, where it shows the point at which people don't care whether or not they are interacting with a machine, and where we will likely substitute away from relating to people - and instead to machines. The indifference curve was based on price, but instead, the "price" we express our preferences at is just from the cognitive load from dealing with this uncanny machine. The "uncanny valley," fits into that because something uncanny imposes a sensory load on us that isn't worth the trouble for what it's doing, so it comes off as weird. When we talk about the Turing Test in a popular sense, this economic curve is what I think we're really discussing, and I think the idea of intelligence and mind is too sophisticated to be effectively reduced to, that a mind is whatever it says it is if you believe it. That little strange loop of logic decouples and unmoors people from reality because once you believe that, you will believe literally anything.