Gigaom AI Minute – February 17

Transcript

Many of the ideas that we have about artificial intelligence come to us from science fiction. One of these is the world of Ex Machina, a 2015 movie, which features a lifelike robot named Ava. A character named Caleb is brought in to talk to Ava, and see if she could pass the Turing test. That is, be indistinguishable in conversation from a human. But the movie quickly becomes more than that as Ava orchestrates an escape from a world in which she is held prisoner. Saying much more than this would give spoilers of the plot away but without doing this, I can say, with regard to the core question of, "Is Ava alive?" and, "Is Ava conscious?" we can never really know.

This is an old problem in philosophy called the "problem of other minds." We don't even know if other humans exist and if they're conscious, let alone if a machine is conscious as well. This puts us in an interesting place, as machines get better at conversing with humans and potentially even pass the Turing test, they will seem ever more alive and ever more conscious, but there'll be no way to know if they are really either. This is an intractable problem, and one can imagine a future, sharply divided on this question, a question that there is no scientific mechanism to resolve.