PodClips Logo
PodClips Logo
Yang Speaks
The Future of Artificial Intelligence
The Future of Artificial Intelligence

The Future of Artificial Intelligence

Yang SpeaksGo to Podcast Page

Andrew Yang, Martin Ford
·
24 Clips
·
Sep 13, 2021
Listen to Clips & Top Moments
Episode Summary
Episode Transcript
0:14
It is my pleasure and privilege to welcome back to Yang, speaks the author of rule of the robots. One of the preeminent authorities on artificial intelligence, and its impact on us and the
0:30
enemy. The author also of Rise of the robots, which was a profoundly influential book on me and the Yang Yang, Martin Ford. Welcome
0:38
Martin. Hey, thanks for having me. Great to be here.
0:42
No problem. I am pumped to have you, and this book of yours, rule of the robots is coming out this week, and I couldn't put it down because there was so much learning and information here, but this is not your first rodeo. So, rise of the robots came out, in 2015, to great acclaim. I think it was a
1:00
Wall Street Journal book of the year, it was a big influence on me and my subsequent works. And for people who are listening to this, who read my book, the war on normal people. You probably recognize Martin's name and lessons because I reference him very very heavily in the war on normal people. So what was the reception to the rise of the robots? You became this very sought-after speaker from corporations and
1:30
Governments, am I
1:30
right? Yeah, I traveled to at least 30 countries. I've gone to a lot of events and still still do. So. It's definitely an issue that resonates with a lot of people. I think that book was fairly well time to maybe you know, the concern was kind of coming to the Forefront. It was always been a little bit of a Divergence between I think economists. Who are they tend to be skeptical of this whole idea that automation is going to be a real problem. I don't know what everyone else.
1:58
You know, you talk to the engineers even technology, Executives people running companies. They're pretty much on board with this. They understand that this is an issue. And of course, then you talk to average people and they see this happening pretty much all around them and and they're on board with that too. So, you know, there are different groups. I think that sort of received it in different ways. But clearly, this is an issue that continues to resonate and, you know, it's the unpredictable. No one predicted for example, that we'd be
2:28
In the middle of this pandemic and the impact that that would have. But I think that trajectory of this is pretty Relentless and, you know, this is going to continue to be an issue.
2:38
I'll tell you more, and even throughout the course of my campaign. I think the disc or shifted, where people started off being very, very dubious. When I made a similar case. Based in part upon your findings about Technology's impact on the labor market, but the conversation has shifted now, to, what do we do about it? I think most people have stopped resisting it and
3:05
To your point. Technologists will have believed this for quite some time in part. Thanks to you Executives, believe this? Because they see in their own organizations. The people that have been on the outside looking in on this conversation are economists. Who've just been Plumbing the data and what are the primary objections that you heard from Economist? And how do you address
3:25
them? Well, I mean the greatest objection that you hear again and again is that this issue has come up many times in the past, right? And we've been through technological
3:35
Transitions, I mean that may be the most dramatic one was when agriculture mechanized, right? I mean used to be back in maybe the late eighteen, hundreds of, at least half of the people in the United States were working on farms and then you had tractors and combines and and all this agricultural equipment come along and those jobs basically vaporized, right? They disappeared and Economist will. Look at that and I say, look, millions and millions of jobs disappeared, but people obviously found other things to do, right?
4:05
Right? And and people in fact are better off. Now. People are happier doing the jobs. They're doing now which require more skill and more education and pay more than you know, being a farmhand, right? So it's all good. And yes, that's definitely something that we could acknowledge but I do think that the transition we're facing this time is quite different. What we've seen historically is that sectors of the economy of kind of automated, you know on a sector-by-sector basis. So first it was Agriculture and people moved into factories.
4:35
Right. So think of the 1950s. We had an industrial economy. Everyone's working in factories. Obviously. Now, a lot of automation has come to those environments, the factories of automated, they vote for also offshored. And so now there are a lot less factory jobs. And what happened is that people moved from factories to the service sector. So now, we're in a world where probably 80 percent of the workforce at least is engaged in service Industries. And clearly, what's going to happen with artificial intelligence, this time.
5:05
Around is that pretty much everything is going to be impacted because AI is really a general purpose. Technology, right? I think it's going to be ultimately almost like electricity. So it's going to invade everything. Doesn't matter. What sector it is. Manufacturing, agriculture, Services, White Collar, jobs, government jobs is coming for, literally everything. So what that means? Is that
5:26
all that, well, the government jobs will be the last to go,
5:28
right? I'm sure that there's some extent, you know, I mean, you know, they've gotten rid of all.
5:35
Tow operators already, right? I mean, I mean they used to be people in Boots, fucking money. I mean, I guess those people do in fact, work for the government, right? So, you know, there are examples of that happening, but I agree. It certainly not not the most efficient part of, of our society. So it'll take longer. But again, what I see happening, this time, is this going to impact everywhere. So unlike where agriculture mechanized and there was this Rising manufacturing sector to to basically absorb all these workers.
6:05
And I don't really see another sector now and what, what can you think of this going to arise in the next five? Ten, twenty years that is going to create jobs for tens of millions of average people that do fairly routine predictable things. Right? I mean that that I don't think that's that's going to exist and yet that's what we have seen historically. Right. You can imagine people do a routine work in the field in agriculture and doing routine work in a factory and now maybe they're working at Walmart. They're
6:35
In barcodes are doing something relatively routine there. So the sector that people have worked in and you know, the definition of the job has certainly changed. But what hasn't changed is that a large percentage of our population may be at least half are doing things that are on some level fundamentally predictable routine. They're doing the same kinds of things again and again and those jobs are basically going to evaporate across the board. So we need to come up with a way to address that we have to think about what we're going to do.
7:05
Do in this new world where those kinds of jobs basically
7:08
disappear.
7:10
And one point I made on the trail Martin was that? Even if you harken back to the industrialization of Agriculture, there were massive social problems and even violence associated with that period. We just celebrated Labor Day. Labor Day was inaugurated as a Federal holiday in 1894 because of mass riots that killed dozens of people and caused the equivalent of today, would be hundreds of millions of dollars worth of property damage. So even if people like, oh, we've been through this before you're like, wait a minute.
7:40
It the big times you're talking about. We're actually really tumultuous. And this is objectively. Going to be much faster and more widespread and more dramatic given the number of Industries impacted and the size of the economy today. The other big objection. I hear from economists is that they think that productivity should be increasing if we were truly on the cusp of automating away, all these jobs. Have you heard that one of a condom is raised that? Yeah. I mean,
8:04
that's the number one, push back. So the issue with productivity what?
8:10
What you hear is that if in fact, the robots are coming online and doing more work than productivity, should be soaring. It should be astronomical. And the reason is the productivity is basically a ratio, right? And the numerator is output. The amount of output. Do you produce? And the denominator is the number of hours worked throughout the economy? That's that's productivity. So if the denominator the number on the bottom, there is going toward zero as robots. Consume all the hours. Then, you know, you basically
8:40
The almost dividing by 0. Right that the number under, you know, the the ratio is going to explode, right? And and in fact, we're not seeing that in fact the last you know, roughly decade or so has been very modest productivity. And you know, there are a number of theories about that. The most common theory is that there is a lag right? That it takes time to assimilate new technologies that? Yes, we're having all this new stuff. Like deep learning. That's finally emerging finally being put into practice, but it's going to take time before business.
9:10
It is really, you know, figure out how to use those in productive ways and it increases productivity. And in fact, that's what we saw historically with electricity. I mean, when electricity was first introduced, it didn't cause a takeoff in the economy at to actually decades, before we figure out how to use electricity in factories and ineffective way. So I think that's a viable Theory. The other thing that I always point out again is that productivity is a ratio and there's a numerator and a denominator and economists, always.
9:39
It's on that denominator. They always say the labor, if the labor goes away then productivity is going to soar but the numerator is output, right? It's the amount of stuff that we produce any economy. And the basic reality is that, you know, no one is going to continue to produce anything. Whether it's widgets in a factory or certainly whether it's a service, unless there's someone there prepared to purchase that right there has to be demand. And so if what we're seeing is that you know,
10:09
Ocean the impact of AI the inequality caused by AI. The fact that that wages are stagnant. If all of that is depressing demand, so that there are less consumers out there that are eager to buy things than that, of course, depresses output. So, so that is one, I think very important idea to keep in mind, and it might be a reason that productivity is not actually storing. So, I think what economists need to think more about is the question. Is there a scenario where? Yes, the
10:40
Bots and AI is actually showing up and having a big impact on the job market, but it's not necessarily being reflected in that cartoonish. Obvious way that they might expect in terms of productivity and I think that they're definitely arguments that you can make there.
11:05
If you're like me or like most people, you probably don't eat enough vegetables because it's covid you're ordering in or your cooking Foods. You like or you forget or let's be honest, like, proactively eating peas isn't and it's not part of my Mo every day. So that's why this episode Yang speaks is brought to you by athletic greens because it's awesome powder. You can throw in some water or throw into a smoothie and get all the vegetables and vitamins that you're supposed to have in your diet. Every
11:35
All day. So one tasty, scoop of athletic greens contain 75 minerals and vitamins and Whole Food Source ingredients, including a multivitamin, multi-mineral probiotic, green, superfood blend and more. And they all work together to fill in all the nutritional gaps, in your diet, increase your energy and focus and a bit digestion. I love this stuff. It helps a healthy immune system and all that. They need to take lots of different multiple products and vitamins and things like that. So if you order right now, we're going to give our audience a free, one-year supply of vitamin D and 5.
12:05
Free travel packs, if you get athletic greens today, so simply visit athletic greens.com, Yang enjoyed Health, experts, athletes, and health, conscious go-getters around the world, who make a Daily Commitment to their health everyday, again, simply visit athletic greens.com Yang. Get your free year supply of vitamin D and five free travel packs today.
12:34
One of the, the
12:35
things I pointed out was that if you have a society where people have no choice but to work and it's inexpensive to put them to work. Then you have a lot of people working in very low productivity areas. And
12:49
so
12:50
yeah, so I think a lot of that too, depressing productivity.
12:53
Yeah, absolutely. And and, you know, some of the Technologies we've created recently.
12:59
Facebook comes to mind are probably not productivity-enhancing, right? They probably have the opposite impact. They actually, you know, cause people to waste more time in the office and become less productive. So it's a messy noisy scenario, you know, it's not something that's just clear cut, but I will you and I think
13:19
while you were doing and do speaking to her for Rise of the robots, which was at Ted very, very high order stuff Martin, and I'm not sure if I,
13:28
Hold you at 10 that I was running for president. No. No, I don't
13:32
think so too. Quite a bit later several months. I so I found out when I saw the article in the New York Times.
13:41
And I must know how to bet. You didn't know I had it in my head. So you and I were talking have to say you and I see things very, very similarly. Now, if you talk to techno Optimist, some of them believe that people will magically rearrange themselves at. Ted, you would I share this where I was, where you always trying to be even-handed and balance. So you were like, hey, FYI, I think this is going to cause some pretty major problems. That some of the people there.
14:10
Well, I guess you know, it'll be okay because they're very, very optimistic, but I came to you afterwards. I was like, yeah, I think they're going to be massive problems. Yeah, I mean,
14:19
Ted is is not just optimistic. It's very, let's be honest. It's very elitist, right, you know, it attracts people who did they're doing extraordinarily well, in this new economy and maybe for some of them is kind of hard for them to imagine that there are people out there that are not sharing in that kind of prosperity. Right? But the reality, you know, the people that go to Teddy's just that uh,
14:40
tiny tiny minority of our population, right? The vast majority of people are going to face real challenges here.
14:50
And you advocated for a version of universal basic income in your 2015 book, rise of the robots. You said, look, this stuff's inevitable. I think you even advocated for it in lights in the tunnel. When did you first
15:02
hear? I actually advocated for it and lights of the tunnel. And yet at that time. I wasn't really familiar with Ubi as a movement and any amount of energy that's been put behind it. So I didn't use the words, basic income, or you bi in that book. I just said, people need an
15:18
Come stream, like, we need to give a money. So so I hadn't quite quite caught up with the lingo there, but by Rise of the robots, I realized that there was actually, you know, people did written about this in the past and there was a movement. And so I think that Ubi is a very important idea. I think that, you know, one thing I do in this latest book, rule of the robots is I have a section where I compare it to the other idea. I hear a lot about which is that the government should guarantee jobs to everyone, right? And
15:48
And that's something that superficially, maybe sounds like a good idea. But if you really sit down and think about it, I mean, it's going to be just massively inefficient. It's not going to reach many people. You're going to have to have this huge bureaucracy, to make sure people show up and do their government job. You're going to attract people away from the private sector to work in government to do what are probably going to be, you know, bullshit jobs. And then you're going to run into all kinds of problems. If people don't show up on time or they have other disciplinary issues and stuff like that.
16:18
So you don't think various solutions to this problem, but I think at the end of the day, you bi is the one that seems most efficient, easiest and definitely, it's the one that is going to reach the most people, write to help the most people and particular help. The people that need help the most, which we could start with the people that are living on the street now, right? They clearly should be pretty close to the top of the list for people that are going to need assistance. So overall, I think that it's a very, very important idea for the future.
16:48
So that we can sort of work on and we find going forward.
16:52
Well, I, you know, I agree, and I think that there are would be real problems with a federal jobs guarantee that a lot of people Overlook until you dig into the nuts and bolts of it. You're like, wait a minute, what happens when the person shows up, and they decide they don't like the job. What? If someone decides, they're not doing a good job? Then, like, what do you do with this person? And in the context of a guarantee? One joke, I told on the trail was like, why not give
17:18
Everyone overall is while you're at it it was a joke, but that, but I think if you have a choice, you'd much rather give people a degree of autonomy and stimulate independent job growth and Entrepreneurship and creativity. Then you would have a new massive bureaucracy that's going to oversee. What's deemed appropriately.
17:44
But yeah, I think that's right. I think I actually, I think your branding of it as
17:48
As a freedom dividend was really that. That was good. That was good marketing. Well, thank you. I think was born of Dick. Yeah, but you're right. I think that one, what do you know there? Definitely are studies that show if you give people a basic safety net, right? So they don't have to worry so much that they're going to literally be on the street if they lose their job or if they quit their job, then they're going to be more willing to take risks, right? And you could probably see more people. Quit that dead-end job and start a small business, you know, and they would know.
18:18
It even if it didn't work out, it wasn't successful the the you know, they'd still still be okay on some level, right? So you would see a lot more entrepreneurial activity in the economy. So I think there are a lot of reasons to you to really recommend the basic income.
18:35
Well, you've been one of the people
18:37
that are that have been getting people to open their eyes. And now in your new book, rule of the robots, you talk about artificial intelligence as the new electricity and then you try and dig into the reality of the fields. That you think AI is going to have a huge impact in relatively quickly and then some areas that it's going to take a bit more time. So first, can you unpack for a moment? This idea of
19:06
Is the new electricity as what you call a gen,
19:08
right? The idea there is that. It's what you would call a systemic technology. In other words. It's not a specialized technology. That's going to impact one part of the economy or one part of society. It's going to transform and and basically enable everything much in the way that the electricity does. I mean if you think about your daily life.
19:29
Virtually everything that you do, to some extent is enabled by electricity, right? If you didn't have electricity, your life would be, you know, unrecognizable and I think that ultimately the same will be true of AI. It's going to be delivered to every industry, every sector of the economy. So it's going to impact Us in terms of, you know, commercially and economically is going to have an impact everywhere, but also culturally, I mean, you're also seeing new forms of art, being being done with AI, you're seeing,
20:00
A, i chat Bots that are being used for people with mental health concerns, you know, to help people deal with depression and things like that. It's clearly going to have an impact in government. It's going to it's already having an impact in the legal system, in some ways in ways that are concerning. So the point is that there is absolutely no aspect of Our Lives, is not going to be impacted by this and it's going to, can you continue to accelerate? So it's like electricity in terms of its reach.
20:29
Each and its impact, but it's also very different from electricity. In the sense. That electricity is a very stable commodity. Right? I mean, it's the same everywhere hasn't changed much since 1950.
20:45
The new electric
20:47
story with AI is completely different. So you can think of it as a kind of a liquid electricity. That's going to be incredibly Dynamic and unpredictable, and and really is going to drive us ever closer to true.
20:59
Intelligence right. True human level intelligence was something that's kind of on the radar in the future. So it's going to be an enormously consequential. And disruptive technology that I think, you know, when someday somebody writes the history of all this, they're going to say that this is the point where
21:18
The nature of our life on Earth, really began to change in a very dramatic way because of the impact of this
21:24
technology.
21:36
Here on Yang speaks Andrew and I are very very passionate about data security and going online without a VPN like expressvpn is like using your smartphone without a protective case and you all know that guy or gal. It's like you're fine most of the time to drop it once and you're toast and that's what it's like not using a VPN to protect yourself. It doesn't take a lot of technical knowledge to hack. Someone. There's like 12 year olds that can do it with cheap. Hardware hackers can make up to a thousand dollars per person just by selling your personal data on the dark web. So you need to protect yourself and that's why you should use.
22:06
Expressvpn. They've got an encrypted tunnel that creates a secure encrypted tunnel between your device and the internet that hackers can't steal its super secure. It's extremely easy to use. I have it on all of my devices phone computer, Smart TV, all of it. It's fantastic. And frankly. I was just in st. Lucia, and I wanted to get NFL games and I'm and college football games. And I was able to do it through my ESPN that I bought in the United States because my expressvpn helped me be in the u.s. When it comes to getting my data. Because it's super
22:36
Or secure its private, and you can choose wherever in the world. You want to be using expressvpn. So secure, your online data today by visiting expressvpn.com Yang. That's expressvpn exp. Ress VPN.com / Yang and you get an extra three months free of expressvpn, and I promise you is worth it. I love expressvpn. So expressvpn.com. / yang
23:09
So you've interviewed the foremost researchers and you've gone to companies and Industry experts and leaders which parts of the economy? Do you think are going to be impacted most quickly and dramatically and then which should people think? Okay, that's not going to happen for a while and I'll share a joke that I used to tell on the trail, which was it's not like your hairdresser is going to be automated anytime soon. No, like the, it would be so ridiculously hard to make it.
23:39
Word Scissorhands. Robot. Like a lot of Highly variable manual, labor is going to be here to stay, but there are some fields that are going to be transformed pretty quickly. And you've talked to folks in the Forefront, what should people expect in terms of Industry change?
24:00
Just in terms of work, the basic lens that you should look at is through is what's the nature of the job? Is it is something that is fundamentally routine.
24:09
Edible and that doesn't necessarily mean low-skilled. It just means that you're coming to work and you're doing the same kind of thing again. And again, you might be coming to work and cranking out a very elaborate report. You know, you might be very much alike. Probably.
24:23
I was a corporate attorney for five months Martin and a lot of that stuff is very boring and rules-based. There's a running joke in law firms, where the first thing you do on a complicated transaction is find and replace because you have a template of another transaction.
24:39
That's kind of similar to this transaction. It's so you put in the new clients name and you just swap out. That's like the first move you make. So there are a lot more highly paid White Collar jobs, like corporate law and accounting that involve. Yeah. So those jobs are going to be
24:54
heavily impacted and the Dare is still I think this this kind of preconception out there that
25:01
You know, all this is something that's primarily going to impact. Low wage workers are blue collar workers or people that don't have that college degree. And that's just not true. I mean there's a whole range of White Collar jobs that are actually easier to automate. Then, you know, the person making hamburgers in the fast food place because to make the hamburgers, you need an expensive robot. You got to worry about dexterity and hand-eye coordination and I'll and visual perception and things like this. Whereas to automate the white-collar job. That's just really manipulating information.
25:30
Sexual lot easier. But in general, what I see is there's going to be impact a big impact in more routine work environment. So I think a lot about the inside of Amazon warehouses. Okay, and then environment, you've got lots of robots already that are mostly focused on moving things around doing what robots can do. But then you've got a whole lot of people there and you know, lately. Amazon warehouse has been a bright spot in the economy in terms of generating jobs, right, but the people in those environments are doing the things that robots
26:00
Bots cannot yet, do which means that they're using dexterity? Hand-eye coordination to pick items off a shelf and then, you know, pack up an order for a customer because the robots are not able to do that. But that will change. I mean just bit Jeff Bezos spoke at a conference a couple of years ago and he said that we did about 10 years. He expected we would have grasping technology that was you know, roughly On a par with the human being, right? And that that's going to transform the inside of those.
26:31
Houses, I think they're ultimately going to become a lot less labor intensive. And the reason is that they're relatively controlled environments, right? You can you know, Amazon has total control over what they're doing there. They can come, they can bring in new robots that do nothing do new things. They can separate the robots from the people. They can control all the issues there. And I think that will move forward. Pretty rapidly where it's going to take longer is areas like self-driving cars, right? Where, you know, yes, the technology. I absolutely believe it's
27:00
Coming. But, you know, it's pretty clear that, you know, getting 99 percent of the way there is is doable, but it's that last 1% dealing with almost all the edge cases the unpredictable events that construction, the weird weather stuff, the closed roads, the pedestrians, you know, that they're drunk or something or not paying attention. All this stuff is really hard to deal with and so I think it's going to be a while. So one of the implications of this I think that
27:31
Is that a lot of the things that get the most hype? And the most attention may actually under perform and to some extent? That may create a perception that all of this is underwhelming, you know, it's not moving as fast as we thought. But at the same time, the things that are less visible less expected inside an Amazon warehouse behind these closed doors. What's happening with software running on computers and automating white collar jobs, you know, that stuff is going to be moving forward, very rapidly.
28:00
And I think it's going to have a big impact. So this is definitely happening. It's going to unfold, but it might not unfold in exactly the way that a lot of people expect, you know, and you've got to kind of separate the hype from the reality. The reality is very real. It's definitely happening, but there is hype. I mean and not to pick too much on Elon Musk, but he is really, you know, into the hype lately. I mean just the other day he announced a humanoid robot that Tesla is working on and he said that a year from now and have a
28:30
Yeah, just for buddy
28:32
said and be able to say to it, go to the grocery store and get me these items, you know, and robot will do that. I mean, you know, I mean Boston Dynamics is a company has been working for years and years and years on these robots. And they finally gotten pretty good at getting them to walk, basically, right? Or even dance. It's fine, but the robots don't do anything else, you know, they don't have any intelligence. They can't go to the store and buy stuff and bring it back to you. Right now. All of that stuff is very far in the future. I think.
29:00
Ink. So, you know, if there's a lot of hype out there to kind of raises expectations around things that we probably are not going to see for a long time. We're not going to see C-3PO from Star Wars, you know, for a long time, you know, we definitely are going to see some very impactful Technologies and offices and warehouses and factories and fast food restaurants and retail stores that are really going to have a big
29:24
impact.
29:36
I know you're all very concerned.
29:39
That we had stopped doing today ads.
29:42
Not, we're still doing it because Yang speaks is proudly sponsored by tushy the modern bidet company with washes away. You know, all that good. Good. When you're in the when you're in the John if you will Weber's poor, John got named after the toilet, whatever. There's probably some history there. That's not the point. The point is that the days are awesome. They save water. They save they help the environment, right? You're using less paper. Less litter less trash. They're cleaner. They're better. They're faster. Many ways.
30:11
Tushy is the modern to bidet for people in the 21st century. Frankly. So just poop wash pat dry, easy easy breezy. And so it simply you've got it uses water. Instead of toilet paper. It's less irritating. It's easy to install it. Attaches your toilet under 10 minutes. No electricity. No Plumbing. It's awesome. Eco-friendly, and stylish better for the world to she has a full product line to make and I quote the restroom, the best room. You've got an ottoman ottoman. Yes.
30:41
How you say it which is a toilet stool designed to help you be at a better angle. You've got the brush that helps you keep your bday super clean and you've got their awesome. Bday that attach you to it. So here's the deal. Start washing with Toshiba day for a better clean. Go to hell. Otoshi.com / yein, get 10% off plus free shipping. It's a big deal. 10% off free. Shipping is a special offer for our Yang speaks listeners at. Hello Toshi.com, / Yang for 10% off. After you buy install, your tushie and show
31:11
Off tag them at hello tissue One Word on Instagram. Please tag on that's hilarious and tag has to because I would love to see that. Hello Toshi. Check it out. Get a bidet. It's awesome.
31:30
One of the things that's going to hold back, self driving trucks is our infrastructure. And I was positing that if you really wanted to give self, driving trucks, and go nationally, you would probably need to just build out a highway expressly for self driving trucks, which would then have very clear sensors and markers. Because one of the problems is having random cars and trucks, interacting with yours like you
32:00
And even if you're 99.9% accurate, you know, you can't have 0.1% crashes. Things become really problematic. And I agree with you that it was fascinating reading in your book about some of the companies that have been working on that problem because I interacted with some of those companies when I was running for president. And the reason why I was so hyped was in large part because of the money involved you're talking about.
32:30
Tens. Hundreds of billions of dollars a year, millions of workers.
32:35
One of the area's you think that AI is going to play a very dramatic role very quickly is Health Care. Can you talk a little bit about that
32:44
some areas of Healthcare in particular where you're dealing with information. For example, you're making a diagnosis. You're doing Radiology right? Looking at Medical images is the one area where there's really an enormous amount of impact right now. So those kinds of roles are definitely going to see increased automation, what you're not going to see. So much in health care is the
33:05
The replaces a nurse right for exactly the same reasons that we've been talking about. She's really hard to build that. So the areas where I see a big impact, our Radiology looking at Medical images, making diagnosis analyzing data in hospitals, so that you can reduce medical errors, you know, there's all kinds of data out there that is, is well, suited to analysis by artificial intelligence and that can have a huge impact to mean before covid came along. I mean, the most one of the biggest killers of people
33:35
Was actually mistakes made in hospitals that were those want, you know, at the top of the list of things that actually causes death in the United States, but the other area related to that, that's going to be critically. Important is is generally scientific research and Innovation. And that of course, will include a lot of things in the medical Arena. Particularly discovering new drugs, new Pharmaceuticals Theodore a lot of companies working in that area, because it's something that's very
34:05
Suited, you know, the artificial intelligence is essentially a form of search because, you know, molecules have a geometry to them a physical geometry. And you can by looking at that geometry and searching for molecules that have a particular configuration. You can actually, you know, find buying drugs, don't have a particular function. So,
34:29
you know, I think that's very exciting. The most exciting thing that's happened in that Arena recently was, was the announcement of deep mines, Alpha Foods Alpha phone system, right? That does fingers out protein folding, you know, the way that protein molecules fold into a geometric shape is critically important because that determines their function. And and this is something that scientists have worked on for at least 50 years. They devoted entire careers to figure out how these molecules.
34:59
Swarm into these various shapes, you know, an easy inexpensive way to do that rather than the very expensive laboratory techniques. And, you know, deepmind managed to basically solve that problem. And so, that's, I think going to be quite revolutionary in the fields of science and medicine biochemistry. So that's the real problem, is the artificial intelligence that it's gonna
35:21
Really kind of augment accelerate amplify our intelligence, our ability to innovate our ability to create new Solutions and to solve problems. And that's something that we critically need because we got a whole bunch of problems, right?
35:37
We do have whole bunch of problems. One of the things that that you mentioned in the book that I found fascinating, was that you thought that a, I was going to accelerate the invention of new materials, where you might have a piece of furniture that like that.
35:51
The small bag and then you can go screwing in the head that it expands. I was like, that sounds like magic. So that's a very and the fact is a lot of people right now, I think are becoming more and more concerned about some of the problems attend with AI but imagining fantastic new materials, may be really enthusiastic. Can you talk about some of those types of
36:18
possibilities? Yeah. I mean, I think it is going to
36:21
accelerate, you know, science and engineering across the board. Right? What you're referring to again. It is technologies that can search for molecules and there was a team. I think it was in the Netherlands. The use this technique to find a new molecule that had structural properties. That would be very strong, very resilient but also could be folded up, very, you know, very tightly. So they, yeah, they were imagining, you know, a bicycle you could fold up and put in your pocket or something like that. I mean, obviously, that's
36:51
Still science fiction at this point. That's the kind of potential that we might see by utilizing these Technologies and the same thing. When we apply this to climate change. I mean, we need Innovation across the board, right? We need not just new sources of clean energy. That's not enough. We need to transform agriculture. We need new ways to make cement. That's not carbon-intensive. We need new ways to, you know, to, to make buildings more to
37:21
Of energy and so forth. I mean, we have a whole bunch of problems. We need to solve if we want to address climate change, not just to mitigate warming but also to adapt to it because we know know some, some warming is locked in. So there are a whole bunch of problems. We need to solve. And, you know, it's interesting that if you look at the technological progress, we've made over the last half century or even the first the last 70 years. What we've seen is this dramatic progress in computers and Communications.
37:51
But overall things have not, you know, we're not getting too into the Star Trek stuff here. I mean we the airplanes we have are pretty much the same as the airplanes we had in the 1970s. The cars are pretty much the same. I mean, they're better, they're more fuel-efficient, but you know, they're not dramatically different. So we've seen we've seen a lot of kind of iterative progress, but we're you know many people like for example The Economist Tyler Cowen is talked a lot about this. How we've been on this technological.
38:20
Jackal Plateau, a kind of stagnation across the board with the exception of what's happening with computers. And you know, my feeling is that we you know, AI is maybe the technology that's finally going to change that. You know, we're going to amplify our intelligence in a way that allows us to accelerate progress across the board. So the key Advantage, the key thing about AI that I believe in is that it's not just about a i itself. It's about what it can do, everything.
38:50
Anywhere else, you know in this is what for example damus hisab is the CEO of deepmind. He says that the purpose of deepmind is to First solve intelligence and then use that to solve everything else and I think that's maybe a bit grandiose, but it's really what we should be striving for and that's the true potential of artificial intelligence.
39:09
And there's that famous saying we were promised flying cars and instead we got a hundred forty characters.
39:16
Yeah, right. And happy that captures the idea that, you know,
39:20
In some areas, we've gotten a lot of progress and in other areas. Not so much. How you think to you know, when I was a kid, I watched Star Trek all the time. We have the communicator from Star Trek, right? And we've got something along the lines of the tricorder, but all of the other stuff, the spaceships, and the warp drive and all that stuff. We're nowhere close to that. Right? So that's what we've seen. Is that progress has become very focused on this one particular area, and we really need to jumpstart.
39:50
Across the board, you know, we need much more holistic, broad-based
39:55
progress.
40:06
Are you staring at a screen right now? Probably are your eyes getting fried because of it? Probably. It's time for you to get some, Felix gray glasses. One of our sponsors on his podcast. These are the blue light glasses at started it all five years ago. They realize the same way. We know that our eyes were not meant to look at screens all day. I'm looking at a screen right now, as I read this ad which is not good, but with Felix gray, you can get lenses at filter 15x more of the
40:36
Important blue light to protect your eyes and the glasses. Look freaking awesome. Mine. Look, trendy AF little tortoise shell. Pretty awesomeness. I've never been in glasses guy. The only ones are like fit. My face are kind of like the Ray-Ban style and these are a band style but they're clear and they look good. So you've got prescription or non-prescription available. You just need to go to Felix Greg glasses.com Yang to check them out. They're durable, lightweight and comfortable.
41:06
Durable and stylish. So here's the deal non-prescription in prescription are available again. Check them out. Now Felix great glasses, dot-coms Yang. That's FB L IX gra. Why glasses.com Yang get free shipping free returns free exchanges physics. Great. Glasses.com. / yang.
41:32
It sounds like you believe AI is poised to get us off this plateau and get the trajectory heading upward again, which is very exciting. One of the most popular concerns out. There is the idea that AI is going to become super intelligent and maybe not need us so much. The technical term is a GI. It's I artificial general intelligence where you'll have an AI that can think for itself.
42:01
Up and then can improve itself. And as soon as I can improve itself, we can do so infinitely. I personally have been trying to tell people. Look, that's not the concern. I'd have. I, I'm more concerned that you're going to have dexterous robot that replaces, hundreds of thousands of workers like that. Concerns me more than having the super intelligence. Where do you come out on the concern around? Like the Skynet type?
42:31
Scenario
42:33
right, in general. I tend to agree with you. I mean, I don't dismiss that concern entirely but I think as far in the future and I think there are a whole lot of issues around AI that we need to be concerned about right now. So we've talked about a lot about the potential impact on jobs, but there certainly are a whole lot of other things. There. Definitely have been documented cases of AI systems being used in very critical areas, the criminal justice system, screaming, resumes doing loan.
43:01
Applications where they've been biased right against racially biased, and also gender bias and things like this, which is, you know, obviously not acceptable that needs to be addressed now, right? That's not some futuristic thing is happening right now, the maybe the scariest thing of all is the potential for autonomous weapons, right? This is something that militaries are to some extent working on already. We could have weapons that literally can Target and kill people without any
43:31
A human being being in control of that. And there's a risk that eventually if those kinds of weapons are developed, they might fall into the hands of terrorists and so forth. So that's a very real issue. And it's again, that's not science fiction. That's, you know, within the next few years. This is something we need to worry
43:47
about. There's a treaty around that, that at least some countries are willing to say, look, they'll always be a human in the decision to kill. And then there are some countries that are like, yeah. I don't know, maybe, maybe we want our vote.
44:01
Hartsfield to Kill
44:04
Ya, so that that's, that's an initiative in the United Nations and their whole bunch of people working in a, I did a very passionate about that. I mean, they really are terrified of this scenario where the work that they've been doing, could be used in such a way. And as you said, there are some countries that have signed on to a band and some that haven't and the unfortunate reality is that one's the ones that have not signed on to it having to be the United States China and Russia, so
44:33
You know, there's not a lot of progress happening in terms of realistically Banning this technology. So I think we need to and minimum make sure that this stuff doesn't get out.
44:45
And into the hands of terrorists and Rogue actors and that kind of thing. The problem with this technology is that there is a competitive Dynamic. So the reason that the United States doesn't want to ban it entirely or band development of it is the u.s. Is afraid that China or Russia will cheat, right? And Russia and China feel the same way. So that that makes it very hard to put this back in the box. And that's why they're not seeing a lot of progress on it. So, that's a really quite a terrifying scenario. And again, this is not
45:13
The Terminator scenario where these machines wake up and become self-aware and attack us. This is other people utilizing this technology to harm other people, right? And that's that's the real fear here. So, but but to get back to your original question, I do think we need to focus most of our attention on these very real issues that are looming right now, but it is true that, you know, building a machine with human level intelligence has always been the Holy Grail of a. I mean, it's if you talk, you know, I
45:42
Did interview 23 of the smartest people working in this field and they're all tremendously interested in this. They're passionate about it. It's their dream, right? To build. And those it was were
45:55
in Martin's prior book, which in some ways is Prelude to this book rule the robots. So he just published interviews with a bunch of the leading authorities in Ai. And then some of those learnings are here in his new book,
46:10
right? So yeah, the title of that book was architects of
46:12
Intelligence and it was basically just interview. So it's the text of the book is, is the interviews I did with those people. And then when I wrote this book, I, you know, do a lot of insight from from those people because there was are, literally you could call them the line between politics, right? I mean, for of the people I talked to Roy one touring award, which is kind of like the Nobel Prize for computer science. So very, very high level people. But anyway, they are working aggressively.
46:43
On building some day and AI system that has human level intelligence. There's a lot of disagreement over how long it might take that to happen. Some people think is going to happen in five or ten years. That would be quite optimistic. A lot of people think is going to be 20 to 40 years and some people think it might be hundreds of years. So we know it's really quite unpredictable. Even the smartest people with the most knowledge, simply, you know, they don't know how long it's going to take, but
47:12
It is true that once we accomplished that it is possible. The machine could become super intelligent. Right? I mean, it could, the general theory. Is that a machine with at least human-level AI would turn its attention to building smarter versions of itself, right? It would refine its own code. So to speak and then could therefore become smarter and smarter maybe very rapidly. That's what's called an intelligence explosion, right? And and people have two takes on that if you're a curse.
47:42
While you are a techno Optimist, you believe in the so-called Singularity. You think that if that intelligence explosion happens, it's going to have dramatically positive implications because everything's just going to take off and we're going to have all this incredible stuff and we're going to have all these medical breakthroughs and we're going to live forever and all of this, right? But then there are other people look at that. No. Well through to, you know, different tint of glasses, right? And see the risk there, which is that once a machine gets that
48:12
Far beyond us. How do we keep control of it? Even if it doesn't become, you know, malignant and actually, you know, intentionally want to hurt us or something. It could in theory, get away from us and get away from our control. And so there are people working on that. Right there. With the most well-known person is Nick Bostrom, right, who wrote the book super intelligence who actually runs the future of humanity Institute at Oxford. And they think about these issues. And there are a few other small Think Tank like
48:42
Organizations where they have very smart people thinking about this. And I think that's a good thing. That's fine, you know, because maybe 50 years from now, it really will be something to worry about but definitely, you know, I've heard some people I think it was Sam Harris, said his tank top. We need we need to have a Manhattan Project focused on this issue. I think that's Overkill, right? This is not something we want to inject massive government resources or something into right now, because it's very unpredictable.
49:12
Active and in the future, but I think it's fine that some smart people are worried about this scenario and thinking about solutions to that
49:20
again, you and I see this similarly in that. I'm more focused on some of the immediate problems though. I do understand the concern like you said and because the concern is potentially species threatening, then you know, it makes sense to make sense for smart people to be thinking about it, but, you know it, but there are some other things going to happen between now and then. So,
49:43
You go around the world talking to people about the perils and possibilities of AI. We've talked a bit about some of the risks. You see, what are the things that we should be? Most excited about it. Sounds like being able to address climate change. Getting off this technology Plateau potentially improving our health. Like here is these are some of the things that you think could be on the table.
50:06
Yeah. I mean all of that stuff The Innovation, you know, the potential to have nude.
50:12
Drugs. The potential to be much better prepared for the next pandemic. Write one of the things you can do with artificial intelligence is search for treatments to an emerging virus in the future. Right? So, so it will be an important tool for that. Also, you know, we should have the we should look at the positive side of even if the issue that we're both kind of worried about, we see what happens to jobs. You remember that? When we eliminate some mundane boring job and replace that with full automation, if we can do that, then whatever is being
50:42
Who's there? Whether it's a product or a service gets cheaper, right? And more available. So if we can adapt to the problems that we see, in particular to the distribution of income, make sure that people continue to thrive, then it's not a bad thing. It's a good thing. Okay, and this is a solution to Poverty, right? To Global poverty. We want a future of abundance. We want a future where all the things that human beings need to thrive, whether it's material things like food, or
51:12
Or things like education, right? The people need to thrive, those things become cheaper. More available easier to access and they become available to people in the poorest regions of the world. Right? In areas like Africa where we critically need to drive development. That's all great. That's a positive thing. That's not something that anyone should feel negative about, but we do know that as that happens. It's going to redefine the the distribution of income.
51:42
And there's a real possibility of rising inequality and that's where ideas, like a Ubi come in, because we need to address that we need to fix it. You know, it's not that hard. I mean, if instead of the real world economy, all of this were a video game or a software simulation, which is kind of what I that was why the the title of my book the likes of in the tunnel is what it is. That's kind of what I was trying to do there. I thought experiment but if you think about it abstractly and you said what if you have a world where
52:12
You could still produce all the products and services which you wouldn't need people to do that. Then how could you fix things? How could you make it good? And any answer is well, you just tweaked a couple things and you find some other way to get money into the hands of people and let the machines do all the work. And I mean, it's all of a sudden you got to Utopia, right? You've got some people call that automated luxury.
52:37
I mean, communism, I think, but I'm a strong believer in the market economy. Okay, I, you know, I believe that we in order to have progress. We, we have to Leverage The incentives that are built into the market economy. And most important incentive. There is the incentive for Innovation, the incentive to become more efficient to the incentive, to get better at producing things, and to produce new things, right? All of that is tied directly to incentives that are provided.
53:07
Guided by the market, right? So even if the one incentive that we all think a lot about, which is the incentive for each of us to go out and work 8 hours a day, even if the internet that particular incentive becomes less important over time, the other incentives, there are part of the market and economy, especially that incentive to innovate and to build new companies to do new things, that's critically important and we want to make sure that we continue to have a thriving market economy. And I think Ubi is probably the way to do that.
53:37
Also, one reason I love you Martin is that you are at heart, an optimist and a pragmatist. I agree with you that if we tweak a few things, this could be a real path to abundance and that we need to harness a i to get us there Martin Ford, one of the Premier Visionaries of our time, his new book rule of the robots. It's a page-turner. You learn a lot. It's impossible. Not to learn a lot from this man. Thank you so much Martin, so great having you. Yeah great to be here. Thank you, Andrew.
54:07
Ew.
ms