Discussion about this post

User's avatar
Keith Winston's avatar

I always enjoy your thinking.

If you only wish to speak about the AI of today, which is of course the only AI we know so far, we miss the future development. It seems to me that it's hard to be confident that AI doesn't go "all the way", whatever that turns out to mean... In which case it's not at all clear to me that "learning makes us human": I think I'd have to say AI is already "learning", and I'm not sure it won't be able to grieve, if that's useful, and I am sure it will have robots with sensors that will have experiences not unlike our bodies with their sensors. I don't know, and you don't know, but what I do know is that humans have spent a lot of time trying to point at "what makes use human" (opposable thumbs! delayed gratification! saved by God!), and I can't tell it's been all that useful.

For now, but maybe not much longer, I think it might be reasonable to develop a more concrete version of "what makes use human" (51% meat!), since... a lot of technologies are on their way.

Which brings me to the last point: if your audience is intended to be people, reading this, thinking about how to adjust their lives, today or in the near future, then your points are more solid than if we're trying to think/talk about all people, including children growing up in a world that is going to be stunningly different by the time they're teens or young adults.

All that said, I agree entirely with your underlying point, as I see it, which is that there are things we can do to help AI help us, and we should be thinking about them. I think this might be one of the most important things to be working on right now.

All expressed thoughts are my own, including those attributed to you, sorry if I've distorted anything. Thanks again for your thoughtful writing.

1 more comment...

No posts

Ready for more?