Microsoft co-founder Bill Gates sat down with Fareed Zakaria to discuss what an AI-driven future may look like, and how AI can help the developing world.
00:00On Wednesday, the White House announced its plan to make America the world leader in artificial intelligence.
00:06The main way it will seek to do that, it appears, is to scale back on regulation of AI.
00:12I recently had a chance to sit down with Bill Gates to talk about how he sees the present and future of AI.
00:21Bill Gates, pleasure to have you on.
00:23We've been trying to track on this show the progress of AI.
00:27And the place we're now at is everybody talks about AGI, general intelligence.
00:34What's the difference between artificial intelligence and what people are now predicting, which is artificial general intelligence?
00:43You know, people use very different definitions.
00:48And you can say, when is AI so good that something like a telesales job or telesupport job, that just having the AI do that work, that it's way cheaper and more accurate than humans are.
01:05So that's, you know, really looking at the labor substitution piece.
01:10Or you can look at the most created things that humans do, like come up with a new drug that helps with tuberculosis.
01:20And, you know, look at, okay, is AI just helping humans get this done or is it eventually replacing humans?
01:31You know, like people talk about writing code.
01:33Simple coding tasks, AI today, can replace human work.
01:37The most complex coding tasks, it's not able to do yet.
01:42And people in the field disagree.
01:46Is that within the next year or two or is it more like 10 years away?
01:51But AI, you know, AI is improving at a rate that surprises me.
01:57Things like this deep research capability.
01:59You know, a few times a day, I take some complex question and just for fun, I see in the AI does an awfully good job gathering all the materials and bringing it, summarizing what I need to know.
02:19Satya Nadella says 30% of Microsoft's code is now being done by AI.
02:23That seems inevitably to mean, one, you'll hire fewer coders, you'll need fewer of all.
02:31I mean, what does, you know, a paralegal do at a law firm?
02:34They do discovery, which is essentially pattern recognition, which is very easy for AI to do.
02:39Similarly, entry-level people at accounting firm, right?
02:43All that is going to mean white-collar work.
02:45College-educated graduates are going to have a more challenging job environment.
02:51Well, when you improve productivity, there are, you can make more.
02:58And so it shouldn't mean, you know, if you get less productive, that's bad.
03:04And if you get more productive, that's good.
03:05It means you can free up these people to have smaller class size or have longer vacations or, you know, to help do more.
03:16So it's not a bad thing.
03:19The question is, is it come so fast that you don't have time to adjust to it?
03:25And in parallel, you know, the blue-collar work, when the robotic arms start to be decent, which they're not today,
03:34that starts to affect even larger classes of labor.
03:38So this is a profound set of changes.
03:44You know, I'm working with Microsoft, OpenAI on these things.
03:50You know, my lens is let's make sure it gets out in low-income countries to help with their health and education and agriculture.
03:58So finally, what advice would you give to young people who are thinking about how do you enter the world of AI?
04:06What should the skills they develop be?
04:10You know, if you get to very profound levels, you almost get to philosophical questions.
04:17In this transition period, the ability to use these tools is both fun and empowering.
04:26You know, I used to have an advantage that I have very smart people I can call up when I get confused about physics.
04:32But now I actually use deep research.
04:35And then I'll send that answer to my, you know, smart friends and say, hey, did it get it right?
04:41And most of the time they're like, oh, yeah, you didn't need me.
04:45And so you can really learn so much.
04:49And the idea of the tutors that people like Khan Academy are building and how we'll get that out all over the world.
04:57And so embracing this and, you know, tracking it will be very, very important.
05:06That doesn't guarantee that we're not going to have a lot of dislocation.
05:11But I really haven't changed my be curious, read, and use the latest tools recommendations for young people.
05:19Be curious, read, and use the latest tools, in this case, AI.
