Iason Gabriel: Artificial intelligence and moral philosophy
Thoughts in Between: exploring how technology collides with politics, culture and society
00:00
Alignment in the Context of AI?
i wonder whether this a good seg way into talking about alinement, which is something you've written about quite a lot. The idea is essentially that if you're building an ai system that has a capacity to learn and canfill a variety of objectives, you really want to make sure it's doing the thing that kind of fulfils human flourishing or human values. A su a misallined air system could be a very powerful system that pursues the wrong objective. And so in that sense, the value allignment question becomes a question for politics,. iam ideally on terms that are fair and that we all accept.
Transcript
Play full episode
