← New Books Network
New Books Network
David Eliot, "Artificially Intelligent: The Very Human Story of AI" (Aevo UTP, 2025)
01:06:38
Loading summary
Transcript71 lines
- [00:04]
A
When did making plans get this complicated? It's time to streamline with WhatsApp, the secure messaging app that brings the whole group together.
- [00:12]
B
Use polls to settle dinner plans, send.
- [00:15]
A
Event invites and pin messages so no one forgets mom 60th and never miss a meme or milestone.
- [00:21]
B
All protected with end to end encryption.
- [00:24]
A
It's time for WhatsApp message privately with everyone.
- [00:27]
B
Learn more@WhatsApp.com this episode is brought to you by White Claw Surge Nice choice hitting up this podcast. No surprises. You're all about diving into tastes everyone.
- [00:38]
A
In the room can enjoy. Just like White Claw Surge, it's for celebrating those moments when connections have been.
- [00:44]
B
Made and the night's just begun. With bold flavors and 8% alcohol by volume.
- [00:49]
A
Unleash the night. Unleash White Claw Surge. Please drink responsibly.
- [00:53]
B
Hard Seltzer with flavors 8% alcohol by.
- [00:55]
A
Volume White Claw Seltzer Works, Chicago, Illinois this episode is brought to you by Indeed. When your computer breaks, you don't wait for it to magically start working again. You fix the problem. So why wait to hire the people your company desperately needs? Use Indeed's sponsored jobs to hire top talent fast and even better, you only pay for results. There's no need to wait. Speed up your hiring with a $75 sponsored job credit@ Indeed.com podcast terms and conditions apply.
- [01:30]
B
Welcome to the New Books Network welcome to the New Books Network. I'm your host, Gregory McNiff, and I'm excited to be joined by David Elliott, the author of Artificially Intelligent the Very human story of AI. The book was published by EVO University of Toronto Press in the US in October of 2025. David Elliott is a PhD candidate at the University of Ottawa, where he researches the social and political effects of artificial intelligence. He is a member of the Critical Surveillance Studies Lab and his work on AI has been recognized with numerous awards, including the 2022 Pierre Elliott Trudeau Foundation PhD Scholarship. I selected Artificially Intelligent because it offers a human centered history of how our ideas and how we think about machines evolved. Moreover, it's a central reading for anyone curious about the dramatic ways AI will certainly shape our society and the key decisions we need to make now before it's too late. It is a much needed and very timely book and it's very clear the author really knows the topic well and has thought very, very thoroughly about the issues he tackles. Hello David, thank you for joining me today to discuss your book.
- [02:39]
A
Thank you so much for having me on, Gregory. I'm so excited to be Here.
- [02:43]
B
Absolutely. David, I want to start with the obvious question, who the book is for and why you wrote it. I do want to quote one or two sentences you give in the book. For example, I'm writing this book so others might see these problems and collectively cry, this is not the future we want. In another place in the book, you call the book a call to Action. Could you talk about these problems and who the audience is here that you want to sort of call to action?
- [03:16]
A
Yeah. So I always say I wrote this book for my friends because I had this feeling where my friends come from every walk of life. They're plumbers, they're marketing executives, they're politicians. And so many people are asking me, what book should I read about AI, where should I start? And I kind of had this issue that the more I started looking at it for them, the more I had a hard time understanding where people should start. Because there's lots of great books on AI out there. Do not get me wrong, I have some of my favorites. But one of the big things was I found that a lot of the general books, the books that covered AI as a general topic, came at it from this fear mongering perspective or from this incredibly optimistic, we're going into this amazing future that's going to lift all of us out of all of our troubles. And I didn't feel like either of those really gave truth to what was going on, or also I felt like a lot of them didn't really speak to the technical attributes of AI and how it worked and give that information. So I wrote this book for anybody who has heard about AI, for anybody who's been hearing all these words in the news or had the bar room conversations, or is scared and wants to know more and wants an accessible place to start that will help them understand the technology, help them understand their place in the world, understand their place in this conversation. And I really wanted to be something that wasn't intimidating, that wasn't a 400 pager, that didn't use heavy academic language, that was just accessible and that would be fun to read. So I say I wrote this book for my friends and I wrote it for anybody who wants to learn more.
- [04:50]
B
Yeah, no, I think you nailed it. And you know, from my perspective, having just read the book, it certainly comes across as measured as opposed to alarmist. I would use fun. Interesting. But I might also say sobering in that you clearly lay out some problems and I would say some decisions that we need to make or at least start having conversations about. And I I want to unpack that, particularly in Act 3 of the book, where you point out, particularly where AI can have a profound influence on society, how we make legal decisions, how we assess and evaluate employees, including teachers, and seems like we need to have some type of framework or reasonable expectations for AI's role there. But let's start at the beginning. And I think the beginning for AI starts with algorithm. And maybe you could talk a little bit about the birth of algorithm and Alcorzini, if I have his name right. Yeah.
- [05:44]
A
Oh, I never know how to pronounce his name either. Say Persian mathematician. I thought it was pronounced Al Khwarizmi.
- [05:51]
B
Oh, perfect.
- [05:52]
A
I have no clue either. I'm definitely butchering it as well. I've written and read it in my head a million times. But yeah, so he's the man who invented the algorithm. And it's so cool because to me, an algorithm, when I first heard this story in my life, was something I heard so much. The TikTok algorithm, the Google News algorithm. We hear that word, but so seldomly do we stop and actually think about, what is this? And I thought his story was a great example of why we make these technologies or the technologies we take for granted. Because when he was working as a mathematician, as you know from the book, we didn't have algorithms. We actually didn't have numbers as they exist today. Yet he eventually introduced that to Europe from India. And basically what he was tasked with doing at that time was figuring out inheritance, because in the Muslim religion, inheritance is incredibly complicated. There are so many moving variables. How many relatives you have, how many debtors you have, the different relationships you have with these people, and it determines what you're supposed to leave behind to each person. And these rules are very important because it's in the Quran, it is religiously important to them. So the, the pressure was high. So he designed a system that anybody could follow, that if they just followed the steps that he wrote down, they could then distribute their inheritance. And this was a very mathematical system. He had to invent a lot of complex math to do it. In fact, he invented algebra to do it. That was the founding of algebra, was his attempt to solve inheritance. But from that we basically got this system that at the end of the day, what an algorithm is, is it's a set of instructions that when followed and when you put the variables in, so your unique variables, you achieve a desired outcome. So when we think about that in a situation like TikTok and the TikTok algorithm, the variables are, you know, the Watch time on different videos we have, what videos we like, what ones we share, and the outcome is keeping us addicted to the app and scrolling longer and it gets fed through this mathematical formula. So I thought it was important to start the book there with his story, because I think that really just feeds into what are we talking about for the rest of this time? What are these devices? And that's all that computing is. Computing is just the mechanization of algorithms. AI is an incredibly complex, living algorithm. So, yeah, that's where I start the book. And that's Al Kirya, Al Karizmi. And it's interesting because his name Latinized, which I've been butchering it this whole time anyway, but in Latin, his name is Algorithmi, which became Algorithm.
- [08:27]
B
Absolutely. And I'm sure there are many high school students who are thrilled with his contribution to their math program. But it is absolutely an incredible inflection point, the development of math. And I was surprised to learn there actually is a Google Islamic Inheritance calculator.
- [08:45]
A
There is.
- [08:47]
B
So you do a nice job. And I should have gone more into this introduction of building on the concepts of themes of AI. And we've just discussed the algorithm and I really, I was very impressed with the next discussion and key characters you tie in. And here I'm going to talk about, or ask you to talk about, Anna Lovelace, Charles Babbage and the Analytical Engine. But candidly, it really starts with Lord Byron, Mary Shelley, and you do such a nice job of. And I don't want to give the book away, so I'm going to let you talk about how much you want to weave those themes together. But in a way, Mary Shelley and her creation plays a role here, which almost seems like the obvious analogy. Yet by the same token, you've got Anna Lovelace and Charl Babbage laying more of the. The intellectual or mathematical framework for AI. So kudos to you on tying those figures together. Could you just maybe go into a little more detail about that?
- [09:39]
A
Yeah. So this was a really fun one. And this is when the. The way I wrote the book really started coming together, because I was trying to figure out how do we do this in a way that's accessible. I knew I wanted to start with that, like, the algorithm story, because I had been told that story and I tell the story in the book of who told it to me, and. And I thought it was brilliant. And then when I got to the Ada Lovelace part, I had forgot at this time who her father was, and I knew I'd heard it before. So when I was doing the research, and I see, oh, her father is Lord Byron. So Ada Lovelace, as we talk about in the book, is the first computer programmer, but she programmed the first computer a hundred years before it was invented, which is just mind blowing and is not given the respect in the history textbooks that she deserves. But her father, Lord Byron, funnily enough, my grandfather was an academic, and during his time, he was the leading Lord Byron scholar in the world. So I grew up just regaled with tales of Lord Byron and around Byron books and just heard all these stories. So to me, that was this almost like beautiful, transcendent moment of realizing Byron's story in place in this story. And then, as I wrote it, just to be a small footnote, these other stories came to bear. So Mary Shelley, the woman who wrote Frankenstein, wrote Frankenstein because of. Well, not because of Lord Byron, but she was convinced to write it, basically by Lord Byron. And I think the story of Frankenstein is the best way to explore AI, so we get to explore that in the book. So these connections weave together in a way that, you know, I'm not a religious person, but it felt very like divine intervention in a way of just like, oh, my goodness, this is. This structure is coming together in a way that it felt like it was the way that this story was meant to be told.
- [11:22]
B
No, absolutely agree. I mean, really, I love the connections there. But please continue, in terms of telling the story, you thought this was a great approach.
- [11:32]
A
Yeah. So then, of course, with Ada Lovelace and Charles Babbage, and I won't go fully into this, so he was this madman mathematician, or a brilliant mathematician, had a side project building the world's first computers. And just. I cannot imagine how insufferable he must have been at the time, because everybody had to listen to him. He was too important not to listen to. But he must have sounded like a madman, just convinced that he could build these things well before the technology was there, well before we had the actual ability to do it. No one else was thinking about it, but he could see the future and he knew what we could make. And the only person who understood him, the only one who fully believed in him, was the daughter of this poet who was herself a mathematical genius. And it's just such an interesting story, getting to learn about the two of them. And as I write in the book, I think the most fascinating part was just that, like, Babbage knew what he was building was important. He knew that it could change things, but he thought what he was building was a calculator. Because what he was upset about was the fact that mathematicians at that level had to rely on other people's formulas to do the grunt work, basically. And he was just upset with other people messing up his formulas and then leading to him taking longer to do stuff. So he just wanted to automate mathematics. ADA saw the future. ADA saw what he didn't and realized that this machine, if he built it, could be programmed to do anything and could become the computer that we know today. And it's. It's so beautiful. I think reading her writing and just seeing that, you know, this young person looking at this thing and knowing what could come if this person was giving the money. And unfortunately, they never managed to build the machine. We know now it would have worked if they had got the funds to build it, but we need to wait about a hundred years until a computer actually got built.
- [13:23]
B
No, that's a great summary. And you hit on so many nice themes there. A, your background, your grandparents, I believe, a classicist and archeologist. I think your parents were comp. Sci professors. And you, you are carving your own career out in academia pretty well. So I want to talk about that perspective, too. Yeah, I think you do a really nice job. Anna sort of saw the vision, and that seems to be a recurring theme. We're obviously going to talk about Turing and Geoffrey Hinton there. And it does seem like there's these individuals who see the vision, but they're waiting for the timing. And I think you do a nice job of tying the pieces in. You know, we'll talk about the. The GPUs and the amount of data and everything where it's almost like generative AI. We've hit the right time, but I'm jumping ahead. I do want to. Yeah, please comment.
- [14:08]
A
If I may comment on that quickly, because I think it's an interesting thing with this book that I tried to really capture is I'm not a fan of what they call big man history, although. Or big person history out of Lovelace, which is obviously not a man here. But I do think it's important to recognize that. I think I love Malcolm Gladwell's book Outliers for this. I think there are so many social forces that create and put us in certain places, but there is often it takes a group or an individual who is the catalyst, who is the person who sees all these things who have come together and makes that action, takes that step, does that change? Or is the one that basically the arms of history almost work through and that's why we tend to tell history in these stories. But as we explore in this book, it's not that these people are necessarily the ones moving it. I think they are the ones who are there to take advantage of it.
- [15:00]
B
Hmm. That's a really interesting point. Could you expand on what you mean by taking advantage of it? Because I. I think reading and just talking to you do think of Anna Lovelace, Charles Babbage, I mentioned Geoffrey Hinton, clearly Alan Turing. They are sort of laying the groundwork for AI mathematically and conceptually. Would you say they're taking advantage of it? Because I almost feel like in our discussion, we're saying more like they're building the framework and advancing it according to the scientific method, so to speak. Yeah.
- [15:31]
A
So, yeah, taking advantage of it might not be the best way to put it, but I'd look at it in two ways here. So one way, I think, like Neil DeGrasse Tyson said something the other day I really liked about how, you know, if Van Gogh hadn't lived, nobody would have painted the Starry Night, but if Newton hadn't lived, someone else would have figured out gravity. They just would have figured it out probably 30 to 40 years later. But those changes make differences in our world, because the time at which we figure it out, the time at which Babbage published that work, did inspire Turing and the work that Turing did. And as we talked about in this book, Turing's influences within his life, you know, being the loss of his first love, affected the way that he thought about the world, affected what he put out, and that changed the course of history. So when I look at it as these people exist in a certain place, but history exists in paradigms, and science exists in paradigms, we can branch off on different stems within science, different research directives. And that's one of the big things this book looks at, is how science actually works, too. Because I think it's a big. It's a really big problem I think, with public education right now, is the way we talk about science and the way we talk about science existing is a linear thing that we progress forward. Where. One of my favorite books is the history of scientific revolutions that we talk about in this book that explores how, like, academia is not a harmonious place, science does not just happen. We don't just build knowledge and build on these discoveries. It is a dog fight. And there are things that we believe that are wrong. We're following incorrect paradigms. Sometimes you have two paradigms going at the same time, fighting against each other. You know, the scientific method is a lot messier than we tend to think it is. Very few people use a very strict scientific method. And like when you talk to engineers and people like that, like I have friends working in computational fluid dynamics, and when you speak to them, they're at the cutting edge of their field. They do not use the scientific method, as we would think of it, whatsoever in their field right now. Or the. My friends who are the researcher there don't. So it's really interesting we started thinking of science in that way and how these decisions that we make shape the branches of history that we go off on. And I find that really empowering because it always reminds us that decisions we make decide where we end up going and that we can make decisions now that will decide the future. We have decide the paths of development we have the things we build. And I imagine later we'll talk about Steve Jobs, because I think in modern time, Steve Jobs is the one who has changed that path the most of anybody.
- [18:10]
B
Hmm. Those are great points. I want to follow up. Would you say, rather than the scientific method, academia, at least with the discussion of AI, it's a little more messy and nonlinear than maybe someone from the outside would expect?
- [18:24]
A
Oh, 100%. It's incredibly messy and non linear. And I mean, as you see in the book, the Second act is mostly about the development of AI to this point, from the time that Alan Turing basically lit the torch and handed it off after his death until now. And you see that it's a very weird path that we took and there's many ways it could have gone in other directions. And you know, it's not always like they were necessarily wrong, like Marvin Minsky's book, which killed the neural network approach and made people think it would never work. I actually read it recently and I read a great piece on it from a mathematician who went over it recently and was like, Minsky wasn't wrong. Minsky was right. With everything he knew at that time and everything we had at that time, he was 100% correct. But he killed this thing off. And as we look at the book, the world changed and the conditions changed. And because the conditions changed, connectionalism and neural networks are functionable now. But yeah, it's a much weirder world. And academia is a very strange place. And building knowledge is a very messy, messy game. It's brilliant. I am a child of academia. I love the university, but I think it's a much different place on the inside than people tend to think it is or than academics like you presented as being.
- [19:49]
B
That's interesting. And I should say you do pull the curtain back a little in the book on academia in general. And as we'll talk about the development of like symbolic AI versus neural network based, it certainly moves in fits and starts. And there's the cold winters. I think, thanks to Mimsky's Rosenblatt's Perceptron. I want to get there. But first of all, just to close out the discussion of Animal Lovelace, she concludes that this machine and a combination of the Jacquard loom, which I think was the predecessor to using punch cards and Babbage's analytical engine, would be eventually very powerful, but never creative. Is that fair? And I want to circle back to that when we talk about.
- [20:32]
A
Yeah, so that's basically exactly what she said in her. Her famous note was that, yeah, it could never be creative. It could only recombine the things that we gave it. And that's the. What we call the Lovelace objection. And that was a big objection against the idea of AI. And in Alan Turing's favorite paper where he or famous paper where he really kind of sets the idea that AI is something that could be possible, he directly argues against Lovelace and says Lovelace is wrong. He doesn't necessarily have the best evidence for his position, but he argues Lovelace is wrong. Computers could be creative and they could make new things.
- [21:10]
B
Yeah, I want to get there. One more follow up for you before we move to ring is the Luddites and the Industrial Revolution. Who were they? What specifically do you think they were protesting against? It wasn't just the advancement or progress of technology per se.
- [21:26]
A
Yeah. So the Luddites are a really interesting group because when you hear them invoked now, it tends to be a very dirty word because the Luddites. Why don't I start with saying how we tend to talk about them now? So the way we tend to remember them now is they were the skilled artisans who saw the new technology that was meant to take their jobs during the Industrial Revolution. So weaving machines, machines such as that, and they were terrified of technology and advancement, so they started smashing the machines and they formed these basically gangs that would smash the new technology, destroy the machines. They were a menace on society. The military had to be sent in to, you know, crack down on them and all that. And that's kind of how they're thought of today is people would say, don't be a Luddite. Don't be afraid of technology. The reality of who they were is much more Complex. The reality is that they were these skilled artisans who were working in these factories and these machines started being brought in and they had originally assumed, oh, we're going to be working with these machines. You know, we are skilled, these machines can help us, we can produce more, we can do, you know, more work. And they could see a future where they could work with these machines. But then the factory owners instead just went, no, you're fired, and got rid of the skilled labor and started producing cheaper, lower quality material. Cheaper, lower quality work that they could mass produce, but could make more profits on and distribute it wider. And the Luddites kind of came at that. This in two ways. One is that as artisans and as people who took pride in their work, they felt like the consumer was getting worse results and that it could be a situation in which, you know, we could have cheap mass produce stuff that they could help make better being the skilled workers, but also they could continue making the nicer, higher end stuff and continue to have jobs. So they saw this as they weren't against the technology, they were against the way it was being implemented, and they were against the way that it was being used to descale labor to, as they saw it, give worse quality products and also tear apart their communities. These resulted in communities just being decimated. So they weren't protesting the technology. They were fine with the technology. They were protesting this loss of power, this change of social structure. And as we actually see the results, they were right. Because the industrial revolution, although we now think about the fruits of it, the good side of it, that time, the next 100 years, were not a happy time. They were a time where, yes, there were so many cheap goods being made, but people's communities were shattered. You know, labor hours for, what was it, 16 hours a day, six days a week, to live in, you know, an apartment with probably five other families. Children were working in factories. Health conditions were terrible because of the condition of cities that everybody was pushed into out of the farms. Conditions were really bad because of these decisions. And again, when we talk about potential futures, I think the Luddites were a group that actually, in a sense, did what I'm asking for here. Not necessarily. I'm not telling us to go and smash equipment, but they dared to imagine a different world with this technology. A world where we used it to lift up their communities and lift up the artisans, instead of a world that focused on, you know, making the maximum profit for the factory owner. And Doug Limu and I always tell.
- [24:43]
B
You to customize your car insurance and.
- [24:45]
A
Save hundreds with Liberty Mutual. But now we want you to feel it. Cue the emu music. Limu. Save yourself money today. Increase your wealth. Customize and save we.
- [25:00]
B
That may have.
- [25:01]
A
Been too much feeling. Only pay for what you need@libertymutual.com Liberty Liberty Liberty Liberty Savings Very unwritten by Liberty Mutual Insurance Company and affiliates. Excludes Massachusetts.
- [25:11]
B
Upgrade your laundry routine with a durable and reliable Maytag laundry pair at Lowe's.
- [25:17]
A
Like the new Maytag washer and dryer with performance enhanced stain fighting power designed to cut through serious dirt and grime.
- [25:25]
B
And what's great is this laundry pair is in stock and ready for delivery when you need it the most.
- [25:30]
A
Don't miss out.
- [25:31]
B
Shop Maytag in store or online today at Lowe's.
- [25:40]
A
Running a business comes with a lot of what ifs, but luckily there's a simple answer to them. Shopify. It's the commerce platform behind millions of businesses including Thrive Cosmetics and Momofuku. And it'll help you with everything you need. From website design and marketing to boosting sales and expanding operations. Shopify can get the job done and make your dream a reality. Turn those what ifs into Sign up for your $1 per month trial@shopify.com specialoffer.
- [26:09]
B
Yeah, it's a great analogy and you circle back towards the end and I do want to talk about that because it raises candidly, it touches on questions about society, the right economic system and what the end goal is. Profit, a beautiful product, the dignity of the worker. But really interesting discussion there. Like I said, I think most of us associate Luddite, as you said, to someone who just stands at the word history saying stop, no more technology. And it's far more complicated than that. Moving to, I think an individual we've already discussed briefly but really deserves a fair amount of time is Alan Turing and his contribution to AI. Could you talk about his background? The Turing machine and the universal computer?
- [26:52]
A
Yeah. So Turing is just, to me, one of the most fascinating figures out there. One of the greatest minds, I think, of all time. And I've been really lucky. I wrote this book two years ago now, and I've been lucky enough that since then I've been doing more work on Turing and getting to read more and more of his philosophy and literature. And I'm so blessed that I've had that opportunity. But yeah, so Turing was a mathematician, but he believed in this idea of building digital computers. At the end of the day, he was the first computer engineer. And in his master's he answered this very complicated Mathematical question through a thought experiment. And the thought experiment was basically a blueprint for what we now call a Turing machine, but it was an imaginary way that a digital computer could work. So he put this out, and although you couldn't actually build the machine, he said it just wouldn't be practical. It basically showed that, oh, we could make this system that just running off of zeros and ones and switches could compute any algorithm that a human could by hand. And that really led to this idea again of, well, if Turing has shown that it might be possible and we could build it, and now we have better technologies than the Babbage period, let's build one. And of course, that got cut short by the war when he went to Bletchley park, was involved in the code breaking. It's kind of this funny misconception now, especially because of the movie about him, the Imitation Game, that people think that what he built at Bletchley park was the first computer. It was not. The bomb is an amazing piece of technology, really cool. Not the first computer. And worked very differently than they showed it in the movie. It's a great movie, but it also is a terrible portrayal of what Turing was like. He was actually a very loved person, very weird, but the people around him loved him. It's like that quirky friend that you have that you're like, we don't understand you. We know you mean no one any harm, and you are way too smart for us and everybody. His contemporaries recognized that he was beyond brilliant. But, yeah, so then after that, he did go on to build some of the first digital computers. Was involved in that. Did not work on the first one, sadly, but he was involved in a lot of other ones. And then from that he ended up philosophizing. And we can talk about this after if you'd like, but this idea of artificial intelligence, of making this philosophical argument, he didn't call it AI at the time. He was talking about thinking machines and made this really strong argument that, yes, there will come a day when machines can think. Here's how it will happen, and here's how we could measure it and tell if it is happening.
- [29:25]
B
I should note you referenced the bomb. I think some World War II scholars feel like it shortened the war and saved millions of lives by a year or two. And I'm referring to his ability to crack the Enigma, which. Unbelievable. You point out, you know, how many different combinations. I'm going to get the zeros wrong. But it was a lot. 150 trillion, please.
- [29:44]
A
Yeah, it was so Much that I had it in the book. And I'm very fortunate. I got to work with historian Kate Lahman Wallace, who came on for about a five month period actually in line by line, checked every historical fact in this book and that was one that shocked me because she got to that and she's like, David, you're missing three zeros. It was already so big. It was an unthinkably large number. And yeah, and the lives he saved, we don't really have a good estimate, but yeah, he saved thousands, millions of lives potentially were saved by Alan Turing. And yeah, we all owe him a debt of gratitude for what he did cracking that code.
- [30:26]
B
Yeah, he really is fascinating individual on so many fronts. I know he's even made contributions to like the mathematics of biological evolution. But I want to talk about a quote you have from him here, just to follow up on what you said regarding a thinking machine. Quote, it need not be a human machine, but rather a machine that can learn so much about humans that under the right conditions it can fool us. We should not think of AI as a human like being, but an algorithm created by humans can act in intelligent ways. And I think that's your paraphrasing of my question for you is does Turing think machines will eventually think and or follow up? Is intelligence a human construct? Because that is kind of a question in the background here. What is intelligence and how do we measure it?
- [31:13]
A
Yeah, and like that is the big question is like, how do we identify what intelligence is? And I really like. So in his famous paper in Mind, he basically says the question of can machines think is irrelevant because we don't know what it means to think or we don't. What does it mean to be intelligent? And I'm really fortunate now to have got to do a lot more reading on what he was working on at this time and get new insights into it that unfortunately do not appear in the book. So I will now share them here, give you the exclusive on this until the next book comes out. But basically what we see at this time was that in the next phase he says we could measure what we would think intelligence is now. And the way we would do that would be a Gallup poll. We'd go around to everybody and say, do you think this machine is intelligent or do you think things are intelligent? Or what is intelligence? But what he was identifying is very much what you said there is. Intelligence is a human construct. Deciding what is intelligent is really a construct of our own imagination. Do we think something is intelligent? It's an Emotional construction. So what he was saying was, right now, I don't think people would see machines as being intelligent. What I am curious about, or what he was saying is, what he was curious about is, will there be a future in which machines can act in a way that it would be acceptable for someone to say, this is intelligent, and not just acceptable for one person because he already thought they were? Would it be acceptable for society as a whole to see the way that these are working and think that they are intelligent? And it's this really interesting thing that I think it's actually terrifying how far ahead he was. Because if we think about the Turing Test. So the idea behind the Turing Test was you had a interrogator and a interrogated subject. You'd have two subjects that they're interrogating. One would be a robot and one would be a human. And the interrogator wouldn't know which was which. And they would have to ask questions. And then the end guess, are they talking to a human or a robot? And basically the robot won if it could render the judgment useless. If the human basically had a 50, 50 shot of guessing which one was a human and which one was a robot, so the robot was fooling them, then we would say, it has now passed the Turing Test. It is convincing this person that it is intelligent. And for a long time, people took that as what Turing was saying. And again, this is one of the tragedies of the situation, is that Turing died very young. And we talk about this in the book. It's a great tragedy what happened to him. But basically what we end up seeing here is we can't really ask him what he actually meant by this anymore. But the interpretation for a long time, which was just built off this paper, was that he was saying the machine is intelligent because it's fooling us into thinking it's human. It's acting like a human. And we call that behavioralism. It's acting like a human. Closer readings of his work have actually shown that that's not what he was saying of that he was saying is that. And this is how I approach it in the book, is that intelligence is an emotional concept and it is intelligent because it is fooling the human. It's not that it's acting like a human. It's intelligent because we think that it is a human. And I think that's a huge, huge difference to really think about there. Because then what I now add on in the new piece I'm working on is this idea that because intelligence is a social concept. And intelligence is something that is universal. It's something that the world has to change to think about it. We have this interesting moment where you realize that what he was actually talking about, based on the last page of this paper and some other papers he wrote before it, is this idea that the Turing Test wasn't meant to say, or might not have been meant to say, oh, when a machine can beat this, that this is what we should use for the measuring stick. And if a machine could beat it, then we would say, ah, machines are intelligent. If you read at the end, it almost sounds like to me what he's saying is, no, if a time comes when machines can do this, society will have switched our definition to say, machines can be intelligent. And we'll start thinking about machines being intelligent, which I find fascinating, because the moment when society started thinking, hey, AI is a real thing, and we actually, outside the AI community, everyday people started talking about it. Politicians started making laws about AI. AI became an acceptable thing to talk about instead of something crazy. Was after chatgpt, after large language models came out that could act in a way that made humans begin to think, maybe machines can be intelligent. And that fascinates me because it's the exact thing that Turing was predicting is that, like, if we could see a machine that could do this, we as a society would shift our idea of intelligence to fit machines within it.
- [35:59]
B
Oh, that is fascinating. And I want to talk about that. Evolution from the algorithm to the deep neural networks and the ability to think, it almost feels like it split the leash. I think at one point you say, we don't even understand how it's getting is and examples of some board games. It actually is coming up with solutions that we would never would have thought of. The last question, because Turing is such a passing individual, could you talk about his notion of the nature of spirit? Because I want to circle back.
- [36:28]
A
Yeah, I think it's a really important factor of Turing that's not talked about that much. And it's one where it is a bit more speculative because, again, we can't ask him and he died before we got too much work on him. So in discussing this in the book, I'm really making assumptions off the writings from him I can find that seem to line up with the other writings that he's talking about technically. But yeah. So Turing lost his first love, a boy named Christopher, right before he went to university and spent his early days at university thinking about the philosophy of the soul and trying to understand if his friend was still with him. And his conclusions are kind of tragic because Turing was like a very practical person. He was looking for these answers for comfort. But being the inquisitive mind he was, he wasn't going to settle upon the comforting answers because that's what he needed. He was going to settle on the answer he found. And the answer that he kind of came to is that there is such a thing as the human soul. There is such a thing as the human spirit that exists separate from our physical body and from the mechanisms of our body, such as the brain, that process these things. But he also made the decision that the spirit had no reason to exist without the body. So although it was a separate thing, and that's what held, you know, our personality, our ability to love, what we might call our humanity, he didn't see it as existing in some other place. He didn't see it as having the ability to exist separate from the body. It needed the body as a vessel. So when the body died, it basically died as well. But in doing that, he created this division that he could say, okay, if the human spirit is one thing, but it's just driving this vehicle. All the things that we think about as being human, you know, the process of power is done by this machine. Intelligence is not, to him, a part of the spirit. Intelligence is part of the machine. So then he later set out to build intelligent machines, machines that could do all these calculations that a human would do, but don't have a spirit to drive it. And that's why we talk about artificial intelligence and artificial consciousness are two completely different things. And I love a nice little tidbit that comes from this that kind of gives support to this. How he was thinking is in his famous paper in Mind, where he kind of sparks the conversation about AI. He's arguing against theological people who say that a machine couldn't be intelligence because intelligence is in the soul. He's like, you know what? I see no reason why, if God wanted to, he couldn't bless one of my machines with a soul. If I build a good enough machine that one day your God could say, you know what? That turn guy is pretty cool. Let's give that thing a soul. And then we could have a machine with both of these. But he's like, in the meantime, I'm happy to build mansions for the soul that will be soulless. Yeah.
- [39:09]
B
That is, again, very thought provoking on how he came to that, how it drove his vision of intelligence. We do need to move on. And as I mentioned, you sort of connect the People and the concepts to get to where we are today with AI and use the analogy of a chessboard. And I'll go from, you know, Alan Turing up to Geoffrey Hinton. Why the chessboard? And could you hit maybe the highlights? You mentioned Minsky, maybe Boolean logic and Rosenblatt briefly. Yeah.
- [39:40]
A
So the chessboard. The chessboard analogy comes from two places. First is that chess is very embedded in the history of AI because there's this really kind of sad thing that the first act was really diverse. We have this Persian mathematician, Ada Lovelace, female mathematician Alan Turing, who is gay. And then suddenly AI becomes an intellectual field based out of Cambridge. And it became so old upper class white. Everybody we talk about for the next eight chapters maybe are these old, rich white dudes. And to old rich white dudes, the pinnacle of intelligence was how good you are at chess. So that was their goal for most of the history of AI was to measure how smart it was by having it play chess. So a lot of the history of AI can be tracked through its ability to play chess, because that's what they were designing it to do. They thought that's what a logical thinking being would do with play chess, which I find so funny. But then to me, chess also just became a really beautiful analogy of. I wanted to use it to talk about the struggle and the fight. And to be honest, I just watched a documentary on chess on Netflix that was strangely, one of the most fun and exciting documentaries I think I've watched in my life. And that metaphor was in my head. And I'm like, I'm going to. I'm going to steal this and talk about the history of AI as a chess game with all these pieces fighting and developing in different ways. And again, got very lucky that the story actually fit the chess game I was describing quite nicely. But, yes, we use that analogy to look at from Turing and Lovelace and then building the board, setting the rules. And then the early people are those early opening moves when anything can really happen. We can go any which direction, but then eventually the rules of the game are set and we end up in this middle game. And that's where we really talk about Geoffrey Hinton, because Geoffrey Hinton is a great example of a good chess player. Your job is to sit back and develop, develop for the end game. Get yourself in a good place to strike at the end. And that's not necessarily a sexy job. It's more of playing smart, playing good. And I'm sure there's also chess players listening right now and saying, this man is not a very good chess player. And they are right. I know very little about how to actually play chess, but yeah. So Hinton really sat back and for a long time was looked at as being part of this school of AI that was not going to work. The connection with school, the neural network school, they were seen as the paradigm that was not going to be the one that was going to create AI that was symbolic. AI was what all the serious people were working on. But I've always loved. It's a quote from Hinton of his master supervisor. His PhD supervisor, sorry, would always say, hinton, I think you need to change projects. I don't think this is going to work. And Hinton would say, just give me six more months, Just give me six more months and it will work. And his supervisor would say, there's not a computer powerful enough to run it and you don't have enough data. He would just say, just give me six more months. And his supervisor was right. He didn't have those things. And it took. I can't remember how many years. I think it was like 25 years or 30 years. I have it in the book somewhere. But the world changed. Hinton did get lucky. The world changed around him. And because of the rise of the Internet, of social media, of surveillance basically, of massive digital surveillance, we produced a lot of data and we gained the type of computing power that we needed in GPU computing to produce the systems that he was dreaming up. And Lee Kun as well. And I talk about Hinton a lot in the book. Hinton was part of only, you know, a big group. But again, I wanted this to be entertaining and not a. Not a technical manual where I'm adding in a thousand different characters. So I, I stuck with Hinton. But like a lot of those guys from the collectionist school at that time are so important to the story as well.
- [43:37]
B
Yep. And I want to circle back that briefly on the. The right. You know, being lucky is better than being smart and getting the timing right. Before we go there, could you just briefly talk about the two schools that are merging in 1950s for the approach to AI? You mentioned Hinton and the neural networks, which is mainly sort of copying the brain. And then you have this symbolic artificial intelligence approach. What exactly is that?
- [44:01]
A
Yeah, so the best way to think about it would be that on the symbolic side, you had researchers that thought that the. The way to create intelligence was to mimic logic. So it was much more philosophical. So it was this idea of giving a machine logical rules. So what we Would call symbols. So describing the world to the machine as symbols. So you might have a set of symbols that describe logically how humans do one thing, or if we see this, this is logically what we would do to that. And it could build up logical rules and then if you introduce a problem to it, it could basically go through those rules and figure out what the answer to the problem should be. So there's no necessarily learning. There are some more complicated pieces to it with decision trees of when it would prune branches on trees. And that's getting way more into the technical side of it. But more than anything, was a much more stagnant kind of program. It was more of a traditional algorithm where the connectionless school is more interested in mimicking the human brain and mimicking, well, how they thought the human brain worked. It was actually mimicked off something slightly different when you do the history of it. But they had this idea of how the human neuron worked, of that we could build mechanical neurons. So the first ones were machines that use switches. And the idea was, is that, you know, you basically train this by having it fail and then giving a reward or a success, and it would change its behavior until eventually the machine figured out what it was doing and, and got good at it. And then that has come into what we have now of these deep learning neural networks, these artificial neurons that are these massive digital structures. They're not physical structures that learn in the same way through a trial and error process.
- [45:38]
B
Yeah, okay, let's go there. Clearly we've talked about timing and Hinton being so patient. Eventually the time came in the form of a deluge of data and GPUs. Tell me why the data's so important and where it's coming from and what makes GPUs perfect for the neural based network approach to AI.
- [45:59]
A
Yeah, so basically with data, the reason it's so important is the way a neural network works is I use the example of taking a kid to, you know, the Humane Society or a pet store to learn about the difference between cats and dogs. So if we took a kid there and they know nothing about what a cat or dog is, might take it, show it a cat, Be like, this is a cat. Show it a dog, say, this is a dog. And then say, I want you to guess which ones of these are cats and dogs. And as you walk through the store, the Humane Society, it's going to start looking at this animal in front of him and be like, oh, that's big. Must be a dog. This one's Small must be a cat. And it's getting pretty good at it. And then eventually he's going to see a Chihuahua and the kid will say, that's a cat. And the parent will be like, no, that's the dog. And the kid might make a rule in their head and be like, okay, well, so dogs can be big or small, but cats are only small. And that's basically how a neural network learns is what we do is we give it, we build a structure for it to learn, and then we give it this data. So for the cat and dog example, we showed all these images of cats and dogs and over time it would start to build its own rules. And that was what makes it so fascinating, is there's no human telling it, you should look for the ear type, you should look for the tail type. It is creating its own rules for deciding what's going to be a cat, what's going to be dog. And then eventually it keeps changing its rules, the weights and the nodes, until it hits a point where with a hundred % accuracy or 99.999% accuracy, with the data we're giving it, it's getting it right every time on the test problems. And then we would freeze that model, say it's done. You could really sit out in the world. But a big problem with it, as I'm sure we're going to talk about later, is if we think about that example I just gave of the pet store of looking at cats and dogs, and it would design that rule that would say dogs can be big or small, cats are only small, because in the pet store it's not going to see a tiger and it's not going to see a jaguar. So if you showed it a photo of a tiger or a jaguar, it would very confidently say it's a dog. Which is why having good diverse data is so important and why we tend to have lots of problems with AI, say, being racist. And one of the big issues with that is in Silicon Valley, where most of them are trained, they were trained from the data in California, the data of the programmers who are overwhelmingly white males. So you had so many instances that weren't included in this data, which led to the bias of this machine because it just never had the opportunity to learn about these things. And then to answer the GPU thing, the best way to think about that is a computer will process, if we think about anything you're processing as a long string of zeros and ones, that's the calculation you have to do. A computer is amazing at processing that super quickly and doing complicated analysis. When we're training these, it's a lot of very little calculations. It's millions upon millions of very, very small calculations to get here. So if you had to line them all up in front of the cpu, it's not using the CPU to the fullest of its potential. And it's going to take a really long time to train. GPUs are graphics processing units which are designed to make the image appear on your computer, to load an image, because an image, it needs to do a bunch of little calculations to show each individual pixel. So it's not as powerful as a cpu, but it's a bunch of basically almost like little computers that help do each calculation at one time. And that is perfect for doing this AI training, because it can run all the calculations at the same time and do it much quicker and much more efficient and much more cost effective. Because computing power is very expensive.
- [49:41]
B
No, like I said, it really seems like the timing finally came together there. And your answer is a perfect segue in Act 3, which you label Frankenstein's monster. And I want my word to talk about who I think we're referring to as the Dr. Frankenstein's of AI, namely Big Tech. And you basically say what we all know, they've got the money, the data, as you just referred, the computing power and the history on their side. And at one point in this third act, you say you have a concern that countries are ceding too much power to these companies developing AI, whose motive is not a better society, but more power and money. How real a concern is that and how do we address that?
- [50:26]
A
Yeah, I think it's a very big concern and it's difficult because I'm friends with a lot of people who work in these companies. And it sound really strange thing of when you talk to these people individually, they have the right idea and, you know, they, they want what's best for society. They don't see themselves as villains. And even a lot of these bigger characters, you know, you hear people like Jeff Bezos or Elon Musk or Mark Zuckerberg talk and they talk from this place of, you know, we want to create a better society for everybody, but we are here to do the good. The problem is that it is just seeding too much power in these individual visions and these very specific visions of what a good society is. And that's not how a democracy is supposed to work. And in history, we have seen that when that happens, things go bad very quickly. When we're allowing an individual person to think for all of us. And also these are people that although those people might have a vision, they're still bound by corporate logic. These are still for profit companies. And for profit companies have different motivation. At the end of the day, the bottom line always has to be the shareholders. That's how companies work nowadays. And the bottom line always has to be, you know, how do we grow, how do we expand, how do we do more? And in this book I use the example of the Smart City in art or the smart city in Barcelona versus the proposed Toronto Smart City, where the proposed Toronto one, which was made by Google, has some really cool tech in it. Like, don't get me wrong, some of that stuff was awesome. But at the end of the day, it always had to come back around to what is Google's vision of the future and technology. And since they always have to have this idea of it being out there and growing, that the tech was very visible, it had some, you know, always a commerce purpose to it and sometimes their vision would align with the human. But in Barcelona, the Smart city's owned by the people and it's operated by the people. If you are in Barcelona, you probably would not realize you are in one of the most technologically advanced cities in the world because you don't see it. They've used it to enhance the living experience. They've used it to do these amazing things like decide where trees should be planted so that you have better climate control in the city, which makes it more walkable and more pleasant to decide where to set up the superblock system. What street should be one way, what street should be two way so that you can have better traffic flow while still maintaining a good pedestrian ability. So that businesses can thrive too. Because what areas will business increase if we have cars? What places will it increase if we have pedestrian? And it's used to make these decisions about making the life better for the citizens within it. And I think it's a really interesting example there of this, this kind of tension. But I think in the book, one of the big things that I try to be weary of is it can be all too easy to say, well, the government's good, we should just let the government have control of it. The people should be in charge. And as I look at here, that side can go bad really quickly as well. So we need to be careful and have this conversation in a nuanced way, I think, of how we want to handle this.
- [53:33]
B
Yeah, that's an excellent answer. And I have to say, as a World War II scholar. I love your callbacks to how private companies and the government handled, let's say, delicate or controversial issues in World War II. I'll let the reader go into that, but that's a great chapter to set up how we are approaching AI today from the private sector and the government. And also that smart city analogy was fantastic. I mean, you even talk about how in Barcelona they're using it to determine where to plant trees. I mean, it really does a nice job of this is how you use AI for the collective good. And you know, in Toronto, obviously I think that's more your home turf, but it sounds like maybe not as, not as altruistic or as holistic in its approach.
- [54:20]
A
We, I think, sorry, just quickly with the Toronto thing, it never got built in the end because there were so many privacy concerns. And that was the thing. At the end of the day, if, if you think about that, a lot of this stuff, when you see this power with Barcelona, they do a really good job of, you know, determining how the data is used. But in Toronto, you start thinking about it, it becomes your daily life. You're the city that you live in. The third places you go to where you have your, your first coffee date is suddenly recorded and processed and handled by a third party company. That's a really difficult thing to, to process. But it is kind of weird because we just accept that now with, I talk about dates there, but everybody who has Hinge or Tinder on their phone is giving all this dating information off to a third party. And now we live in these digital spaces now. So we talk about that in the book as well.
- [55:12]
B
Yeah, let's dive into that because that's a great point. Do you think AIs need for data is increasing the surveillance state and eroding our privacy 100%?
- [55:22]
A
It is. AI is built on a foundation of data and the change in how we collect data too, which is just wild. One of my early academic publications was a warning about this where it was interesting that Google was coming out and being like, you know, we've heard your privacy concerns. We, we heard you. We're going to be better. We're going to get rid of cookies on our browser. And I remember hearing that and like, that's way too nice. You're hiding something. And it was this, this moment of kind of what we were going through was at that time, what was the most valuable data was private data. Was this data about, you know, this individual that we needed to track you personally around and link it to you for that data to be valuable because we were using it for advertising. But now we are moving into this place where just what's valuable is just information. So, like, we're going to follow you around, but, you know, we're not going to write down your name with it, but we will follow you around and track and see what you're doing and, you know, create all this data because we don't care who's doing it. We just needed to build these machines. But that still is a massive development of the surveillance state. And then when we put more and more machines into our daily life, like AI works through surveillance, you think about a robot, you know, now we're talking about these Tesla robots everywhere and, you know, AIs, the meta, smart glasses and all that, all of those are camera operated, all of them have computer vision technology. Like that is digital surveillance that is now going to be working all the time, all around us. And it's just like the more you think about it, the difference between, you know, a human looking around everywhere and, you know, a lot of the time you look at a face passively and you don't even try and recognize it. Yeah. Facial recognition algorithm doesn't know what faces it should or shouldn't try to. And it's never going to turn off. It will just keep going. So, yeah, this expansion of the surveillance state and it's, it's really interesting, the personal surveillance, the more you think about it, that when Google Glass came out in 2012, people were like, we are not okay with this. The general public, it was one of the biggest issues because people got punched for wearing them in public and people ridiculed them. We called them glass holes. Now people are wearing meta glasses everywhere. People are wearing smart glasses all the time. They also look a lot cooler. And nobody has a problem with it because we've just slowly, over the last 15 years, become so used to our privacy being eroded, become so used to our data being collected. And just assuming that, you know, Google and Facebook are always listening and collecting all these things and we joke about that. I don't think they're actually listening to us. But I always remember a Google employee once telling me I was sitting in a bar with him and I went, tell me the truth. Are you listening to me? Are you listening to me through the phone? It's too creepy. And he looked at me and went, no, we are not. But if you knew how we did it, you'd be even more scared. And it's just their algorithms were so good, they were collecting so much Data that their algorithms were so good at predicting what we wanted before, like you, you hear the story, there's the famous story about the target algorithms predicting that women were pregnant before they even knew.
- [58:35]
B
Yeah, it is absolutely amazing. And I think you referenced this earlier, GPT. And I want to get back to that because it has 175 billion parameters. I mean, these machines have. Now, my words slipped the leash. But when you were referring to AlphaZero, once it started learning on its own, we came to paralyzing realization that it was unbeaten and we couldn't even understand how it was playing. You write GPT was doing things, quote, it wasn't trained to do up to GPT. We understood the math, we understood the programming even as we evolved or settled on the neural network approach. But once we come to sort of this deep neural network with multi layers and billions of parameters, this thing literally has a mind of its own that we don't understand. And as you point out, it's being used in very serious use cases for society. I mean, sentencing, bail reform, assessing teachers. My question for you is, is this a concern that we don't understand? And how should we, where should we go at this point? Should we just put a moratorium on AI until we do?
- [59:51]
A
Yeah, I think the big problem with this is there's kind of an assumption by some that we will understand it someday. I'm of the side that we never will. A deep learning system just is too complex to be reverse engineered. It's like trying to think about where a human thought came from by tracing the neurons in the brain. It's. I don't think we're going to be able to do it, but that doesn't mean we should stop using it. The same way we don't stop using humans because we can't always trace their thought process. But it means we need to think about where we're using these systems and when and how and have clear guidelines around it. And we're already doing that. The European Union has the most robust guidelines for this in the world, which I think actually answer some of the major concerns that you have expressed here. Because for them, basically what they do is they say, here's our definition of AI. If you are using AI system, we are now going to make four tiers of risk. I think it's four, it might be three. I need to double check on that. Within each tier of risk, there are different regulations. So if you're using it for stuff in the lowest tier, you know, that'd be like creating art, creating GIFs, stuff like that, they're like, it's harmless, don't worry about it. We're not going to regulate you at all. The second tier, they're like, okay, now we're getting into a place where we need to put some regulations on you. And that might be places where it's making a management decision, it's making a hiring decision, a decision of who's getting a bonus. And at that point they might say something like, okay, you need to be able to audit this. You need a mechanism to go through and see how it's making the decision. So they would have rules like that. The third tier gets more intense. And then the fourth tier is a level where they say, no, you, you cannot touch this. If you are a company working in Europe, you cannot build these systems. These systems are illegal to operate in Europe. That stuff like social credit systems, all of this kind of like, I think autonomous weapons systems are on that list too. Like, you can't build an AI that will make the decision to kill somebody. Which is being done right now in Ukraine. It's the testing zone for autonomous weapon AI. And you know, I have heard I need to go and double check before I start. That's huge authority. But there are systems guardian systems being operated that could make the decision that something is a target and pull the trigger. I was told that I'd need to triple check that because it's a huge claim, but I wouldn't be surprised if we've already written the code for that. It's more of a question of have we put it in place in the battlefield yet? But, yeah, so I really believe there are certain ones that we should say no off limit. But I like the European system of that. It should be a graded mechanism to look at what is the actual risk here. And then what rules do we need for this?
- [62:33]
B
And you end the book actually on an optimistic tone. I think the last chapter is labeled Hope and you quote Tolkien, but you do note that Turing lived between the wars and the brutality from World War I to World War II increased. The industrial revolution had enormous negative effects for employees. What gives you confidence that we'll learn from those lessons and use AI appropriately and not let it, you know, not let it exacerbate the problems of the past.
- [63:06]
A
Yeah. So I think with this, like, the tone of the book is hopeful and I feel like some people criticize me of being too optimistic. And I like to say that optimism and hope are related, but they aren't the same thing. I am a big hockey fan and I am hopeful every year that the Ottawa Senators are going to do good, but I'm not optimistic about it. And I think that's the difference, is that I have hope. Hope is what gets me up every morning. Because when I look at these stories in the book, as we talked about, history moves and it moves based on our decisions. History is not written in stone. We look back at it as this linear line because that is how it ends up being written. But in the moment, we are making history every moment. And I have hope that we can guide ourselves towards an optimistic future. I can see a future that is optimistic. I can see a way that we can implement AI that lifts everybody up, that lifts the bottom. That is the most liberating technology that humans have ever invented. But standing at this point, I can also see a terrible place. I can see a place in which is used in war and place that is used to make more suffering. A place in which it automates our jobs. And we do not put a good social safety net in place. And we end up with a lot of young people being very disenfranchised. And we all know how that tends to go with populism. We've seen that in history before and that is a very real possibility. But I choose to wake up every morning with a. A moniker of hope because I believe in people and I believe that we have the opportunity right now to write the rules. We're living through history, you and me, and we get the opportunity to push this forward. And I really choose to be hopeful. And I think that was the reason I wrote this book for the everyday person of. I want to give the everyday person or anybody who reads this book the opportunity to join this conversation and the opportunity to join this movement of hope and try and push this technology towards what it can be.
- [65:07]
B
Well, David, I hope you're right. I mean, given how much time you've spent looking into these issues. And it's clear from the book you're as knowledgeable on it as anyone. But like I said, it does still feel like there are some very big forces at work here. Regardless, thank you very much for a great interview. Again, the book is Artificially Intelligent. The Very Human Story of AI by David Elliott. David, great interview and thanks for writing such a thought provoking pilbering book.
- [65:36]
A
Thank you so much for having me, Gregory. It's been an absolutely amazing.