Four: Brian Christian on artificial intelligence
Effective Altruism: Ten Global Problems – 80000 Hours
00:00
Is There a Risk of Uncertainty in Machine Decision Making?
In this case, the fact that the bike walker isn't in the data sat literally kills them. And so there's something really kind of haunting, i think, about a model that is powerful enough to essentiallyin force the limits of its own understanding. It has this impoverished view of the world, but then it forces the world to conform to that, rather than itself adapting.
Transcript
Play full episode
