In the film Ex Machina, tech mogul Nathan Bateman (played by Oscar Isaac) invites his employee Caleb (Domhall Gleeson) to his remote compound so that Caleb can perform a Turing test on Ava, a robotic AI that Nathan designed.
On the morning after Caleb meets Ava for the first time, the two men have a very interesting conversation about how to test whether she’s actually conscious.
Caleb: It feels like testing Ava through conversation is kind of a closed loop.Nathan: It’s a closed loop?Caleb: Yeah, like testing a chess computer by only playing chess.Nathan: How else do you test a chess computer?Caleb: Well it depends. You know, I mean you can play it to find out if it makes good moves. But that won’t tell you if it knows that it’s playing chess. And it won’t tell you if it knows what chess is.
Nathan describes the distinction Caleb is drawing as “simulated versus actual.” We could also characterize it as syntax versus semantics. Either way, it gets to a fundamental problem with using linguistic competence to assess a machine’s general intelligence or its capacity for consciousness.
In his 1950 paper “Computing Machinery and Intelligence,” Alan Turing describes a variation of a popular party game of the time called the “imitation game.”
In Turing’s version of the imitation game, a human being passes messages back and forth with two interlocutors in two different rooms. One of the interlocutors is another human. The other is a machine. If the human can’t tell which of the interlocutors is the machine, the machine passes the test. It can be said to think as humans do.
But not everyone agreed that the Turing test was a valid way to demonstrate machine consciousness. For philosopher John Searle, thinking meant not just the ability to speak, but also to understand what one was saying. There’s syntax, which relates to the rules for putting together grammatically correct sentences; and then there’s semantics…