Factory reset: The machines are ready — are we?

In this episode of People B4 Machines, host Amanda Cupido dives into the transformative world of factory automation with guest Adam Dorr, Director of Research at RethinkX. Together, they explore the rapid advancements in artificial intelligence, the rise of humanoid robots, and the challenges and opportunities these technologies bring to industries and society. Adam shares insights on the balance between automation and human adaptability, offering a glimpse into the future of work and production. Tune in for thought-provoking discussions on how we can navigate this revolutionary era while keeping people at the heart of progress.

For more bold questions and sharp insights, visit www.peopleb4machines.com. Remember, the future isn’t fully automated—it’s people-powered.

Transcript

Copy

Adam Dorr 0:00

Intelligence is the solution to every problem. We look all through human history, we look forward into the future. Any problem we've ever solved, any problem we ever could solve, intelligence is sort of the the key.

Intro 0:17

People before machines. Conversations on the chaos of factory automation from Monday morning to the very near future. An eclipse podcast.

Amanda Cupido 0:26

Welcome to People B4 Machines, conversations on the chaos of factory automation from Monday morning to the very near future. I'm Amanda Cupido, a speaker, author, and entrepreneur with a passion for the intersection of technology and humanity. Today we're discussing if we as humans are truly ready for machines. I'm joined by Adam Dorr, the director of research for RethinkX. He is an environmental social scientist and technology theorist whose recent RethinkX publications have focused on the disruption of the global energy sector as well as the implications of energy, transportation, and food disruptions for climate change. Adam is known for speaking about RethinkX's work on stages and in the media. He has more than a decade of teaching, lecturing, and public speaking experience. And I am so glad he is here with us today. Welcome, Adam.

Adam Dorr 1:21

Thanks so much for having me.

Amanda Cupido 1:22

Let's dive right in and start off with talking about the current landscape right now. We have been automating our lives for decades, but what's different right now?

Adam Dorr 1:31

Well, the elephant in the room, as far as automation goes, is the staggering leap forward that we all experienced in practical terms with AI. So we've been hearing about AI for generations, for decades now. And it was always sort of a it's coming, it's coming, it's coming sort of thing. And for a long time, what we referred to as AI in industry and applications was really just advanced software. But with the chat GPT moment in November 2022, several years ago, AI made this leap into feeling like something more out of science fiction. And the reality, of course, is that behind the scenes in development, it was not a completely abrupt process. It was an incremental advancing development. But what it felt like in everybody's lives was wow, this is a brand new thing under the sun. Okay, now in the years that have followed, that artificial intelligence technology has just been on a stratospheric growth trajectory. It is getting better and better and better, and it's becoming an integral part of our lives. Now, for uh for automation, what does all of this spell? Well, it is on the software side, on the intelligence side, machines are now suddenly able to understand and interact with us in a very human way, especially through human language that we've never experienced before. So, yes, automation in production, automation in manufacturing, automation in industry is nothing new. We've been doing that for decades. But what is new is the degree to which software feels like it's genuinely smart. And once we get that artificial intelligence embodied in machines and not just as text in a chat window, then we're going to start feeling a new, a categorically and qualitatively different kind of impact. And so that's the exciting thing to be thinking about that movement from AI, which is just on your screen, to AI that is embodied and able to interact and operate in the physical world with the same remarkable intelligence. And what that is going to do for production across our economy is going to be really quite dramatic. Revolutionary is probably not an exaggeration this time. Normally it is when you hear that word, you know, yeah, maybe with take it with a grain of salt. But this probably is. If anything meets the criteria for being revolutionary, it's going to be the embodied artificial intelligence that we're seeing coming.

Amanda Cupido 4:12

I love the little history that you gave here because I feel like that sometimes doesn't get acknowledged that the word artificial intelligence even has been used since the 50s. But what we're seeing now is just this supercharge. And mainly because of computing power that's been unlocked. And a lot of academics have even been saying that this was possible since the 80s, but we're finally having it come to fruition now. And so as people are trying to wrap their head around this, there's obviously a group of people who knew this was coming and could have predicted it, but everybody else is sort of catching up. Do you feel that people in general are underestimating the impact of this revolution in their professional and personal lives?

Adam Dorr 4:54

I think we're actually doing a bit of both overestimating and underestimating, which is a way of saying we haven't really quite got our arms wrapped around this yet. And we don't have a really clear idea of what to expect. And so some things are going to continue to surprise us and exceed our expectations, and then some things are going to continue to disappoint us a little because they're not going to come quite fast enough or as fast as we might be expecting. So I think it's going to be a bit of both, but certainly some capabilities have arrived more abruptly and sort of to a greater extent than we might have imagined. I'll give you one specific example. We did not expect that AI would become so good so quickly at producing creative things like art, images, you know, images that look like paintings, images that look like very artistic photography, even music, I mean, even songs can be generated by what we call generative AI. The leaps from just, you know, what looked like just scribble, just just very, very primitive capability just a few years ago to now when at the click of a button you can get something that's very convincingly realistic or very stylistic. And it it takes some nuance to understand whether, well, is that really creative or is that derivative? I'm not an artist, but to me, these things can make pretty beautiful pictures. And I mean I'm I'm I'm sure a real artist could tell you, oh, well, that's not so great, et cetera, et cetera. That looks amateurish or whatever. But holy moly, I didn't see this coming. And I've been paying very close attention to this. So that's an example of where we underestimated the capability. I think where we're where we might be overestimating the capability is how quickly we're going to see something similar in making, you know, C3PO commander data from Star Trek and Star Wars, you know, the robots walking around, you know, talking to us and really being able to be capable in all the ways that people are. We're probably overestimating that in the near term. But the general pattern to follow there is we tend to think linearly, not exponentially. And because we tend to think linearly, we overestimate what's going to happen tomorrow and next month and maybe next year, but we still nevertheless underestimate what's going to happen in five or 10 years because of the acceleration of the growth and the improvement. And so even though we may be sort of overhyping these capabilities, if you hear folks saying, well, it's right around the corner, next year there are going to be robots walking around in our factories and putting, you know, factory workers out of jobs, not next year. But that doesn't mean it's going to be 50 years. It's probably more like 10. And that means, you know, 10's a not very long in a lot of the economy. 10 years is sort of the planning horizon for plenty of industries. And so this isn't something we have to pay very close attention to and get ready for. But uh, is there going to be an army of robots marching on us, you know, to replace not next year. No. But at some point, yeah.

Amanda Cupido 7:52

And what does that rollout look like to you? Is it slow and measured with one humanoid and then a second and then a third? Or is this kind of like a supercharged chat GPT moment where we realize these things are great and suddenly half the floor is these humanoids?

Adam Dorr 8:08

It isn't going to be so much of an either or. It's going to be all of the above with automation. So there are things that specialized industrial robots can do today that they're still going to be amazingly optimized and the best option for in 10 years from now, for sure. So definitely going to have continue to have very large, very specialized industrial robots. And they will benefit from increased artificial intelligence operating and driving them. They will become more adaptive, more resilient, less error prone, less brittle in terms of their operation and their operational parameters and so forth. So they will become better. But they will be joined by, not replaced by, but they will be joined by robots of other kinds. And that could be a quite a wide variety of other embodied artificial intelligence. So think robots on wheels as well as robots on legs. I think we're going to see a variety of forms emerge. However, my team has worked through this in some detail, and we do believe we have good reasons to expect that the human form, humanoid robots, will be a winning case for several reasons. I'll list a couple of the big big reasons. The first is that much of our production environments are built right now around the human form, around human scale, human tooling, human needs, and so forth. And so dropping a human-shaped platform in is a good fit. You don't have to do too much modification. So it's a direct, and that means relatively easy, both financially and engineering-wise, a drop-in. Okay, that's one reason. Another reason is that human beings, the human form, is a pretty good general purpose platform already. I mean, nature proved that. We're pretty darn good at a lot of different things. We're pretty adaptable, we're pretty capable. You know, dolphins are smart, but they don't have opposable thumbs and they have a hard time when they're not in the water and so on, right? And, you know, birds can fly and that's great, but you know, they're they they don't have a they don't have opposable digits again, and they can't manipulate tools and so forth. So human form is pretty darn well capable already. Now, is it possible something could be even better? Like a crab or something with eight legs and you know, eight arms and legs instead of just four, like we do, two and two? Perhaps. But we know that human beings work really well. And so a human-shaped machine makes very good sense. And then the last big reason is that if you want to build something at scale and drive its cost down, it makes more sense to do something that you can make a lot of, millions of, and get those benefits of scale, of economies of scale, than lots and lots and lots and lots of specialized forms. And so that would tend you to be looking towards a general purpose platform anyway, for those benefits. I suppose I'll mention one more uh important factor there, which is why general instead of specialized embodied AI? Why a general form? Well, a general form can be repurposed very quickly. You can have the same robot, the same machine, if it's a general purpose machine, you can have it do one set of tasks in the morning and a completely different set of tasks in the afternoon. If you have specialized machines, that sort of skill transferability, the mobility of performance of tasks, or the adaptability as needs require, that's more difficult with specialized designs. So for all of those reasons, a general purpose machine is probably going to be a winner. It'll emerge early, it'll be more affordable more quickly than specialized ones. And so for those reasons and some others, humanoid robots are likely to be an outstanding success here in the near term. In the longer run, we'll have everything, but in the near term.

Amanda Cupido 11:53

So interesting. I can really picture it in my mind. Thanks for giving these examples and digging deeper on examples. I'm wondering if there's an anecdote or story that you could share of either where automation went too far or not far enough that we can have as a cautionary tale now as we look ahead with all of this.

Adam Dorr 12:16

I think it's very difficult to predict exactly how adoption is going to proceed. An enormous amount will depend on the specific use cases. Virtually certain that we will continue to be surprised at which tasks these machines are capable of performing and when. So this is sometimes called known as Morovek's paradox, where things that seem like they should be easy because they're relatively easy for us have proven to be very difficult for machines to do well. Like just getting a humanoid robot to walk with the kind of finesse and balance and grace that a child has has been enormously difficult. And yet, getting artificial intelligence to paint a realistic looking picture in a fraction of a second is has proven easier than what children can do without even thinking. And indeed, what some animals can do an hour after they're born, like a horse can walk an hour after it's born, we are going to continue to be surprised by which tasks prove difficult and stubborn to automation and which ones are relatively low-hanging fruit. So it's not clear yet. However, if I were to hazard a guess on a 10-year time horizon, what does it look like? I think we probably have some industries embracing this very quickly, bringing on not just one at a time, but okay, we're going to do a full, a clean sweep here. And that may work for them, but they'll struggle and discover that that's that's imperfect. We may have others that are slow, that buy one at a time, two at a time, integrate those in with alongside the existing workforce, and that will have its pros and cons. And ultimately, my guess is that every industry will find the balance that serves them best, that maximizes their value. And another thing to think about here is that it's because they because the this technology will be in a massively deflationary regime, it timing will be a tough choice, a tough economic choice. Here's what I mean by that. When do we decide to sort of upgrade to the next laptop or the next phone? It's a tough decision, right? Because you know the next one's gonna be better. But you know if you wait another six months, it's gonna be cheaper. So, you know, when do you dive in? When do you make the leap? Well, boy, that's gonna be an enormous challenge for businesses to say, okay, are the robots capable enough? But if we just wait a year, they'll be half the price. You know, do you buy one or 10 now? And it's so that those sorts of challenges, um, it isn't going to, and I guess my point is, it isn't just going to be about the raw capability of these, but about the the um the value proposition that they represent. Uh that will that will help individual firms determine how rapidly do we onboard these, uh, what schedule, and so forth. It's risky to try to automate too early in your sort of product lifecycle and production lifecycle before you've got all the wrinkles ironed out, before you've got designs finalized, uh, because it's enormously difficult to do retooling and readapt and redesign your automation needs. And it turns out that human workers are pretty darn capable and uh not so easily replaced, even with very capable machines. Now, will production eventually be more automated? Certainly. And will different kinds of robots than just industrial robots, so in other words, will humanoid robots be part of that picture? Yes. Is that right around the corner for the majority of production tasks involved on, for example, an automotive manufacturing floor? Not right around the corner. But again, on a time on a 10-year time horizon, yes. Yes.

Amanda Cupido 15:49

So given stories like that, which I know can instill a lot of fear in leaders and even in employees themselves, what advice would you have for organizations who are trying to help their team embrace change and have a growth mindset about all of this?

Adam Dorr 16:06

Well, there's several things to keep in mind. Uh, the first is that let me be the first to say this also, that nobody has all the answers here. Nobody. This is terra incognita. This is uncharted territory for humanity. We have never faced anything really quite like this before. So 10 years is maybe a bit aggressive, but certainly on a 20-year time horizon. So, you know, imagine 2045. Okay. At that point, somewhere along the line, machines are going to be capable of doing virtually anything that a human being can do, any economic task. Now, we do more much more than just economic tasks, but on a 20-year time horizon, it is very difficult to see how most of what of the work that we do in our modern economies could not be done by machines for a tiny fraction of the cost. We've never seen anything like that before. This is very different than mechanization, where we get large productivity gains and efficiency increases from using machines, but nevertheless, those machines have to be guided by human beings. This is a key difference. So let me be the first to say that nobody has all of the answers here. Nobody. What that means is we need an approach to this challenge that is all about learning. That is the most important thing to onboard here. If we had a playbook to follow, then it would be just a matter of, okay, here's what you need to do and when you need to do it and how you need to do it. But we don't have that. It's making the playbook that's the important task right now. And everybody has a role to play there because learnings can come from every corner of the economy, every corner of the world, frankly. So what do we need, what is our most important job right now? Keep our eyes open, experiment, pilot projects, have a very high tolerance for failure. In other words, know that we're not going to get this right on the first try, that we're going to have to be, you know, zigging and zagging, tacking back and forth, you know, making adjustments, and onboarding those learnings whenever and wherever they come from as rapidly and as with as much agility as possible. Now that is very difficult for some industries, manufacturing being one of them. All of that is a lot easier said than done, especially if you have enormously capex-intensive production, for example. If your industry is one that has to make really, really large bets on processes and equipment, it is very difficult, this prospect of we might have to change a lot next year or the year after. But that's the world we're entering. And so that's the first and most important thing to keep in mind is agility, experimentation, failure tolerance, and learning. The second very important thing that I want to stress is the focus needs to be on protecting people here. And I say that too as a at the societal level. We don't want to be protecting individual firms. In other words, we don't want to be bailing out individual companies. That's not the point. It isn't the companies that matter, it's the people in them that matter. And so whatever we face here, whether there's a lot of displacement of workers or whether there's a lot of you know shakeup in an industry, if we need to keep our eye on what really matters, which is the people of our societies and not the rest of it. Okay, so that's the second very, very important thing. Now, from there to down closer to brass tax, what can companies do? Probably the most actionable, the most feasible thing is to adopt a posture, again, like I said, of experimentation and bring these set up piloting stages, staging domains within your operation. You want to be learning with your own hands-on experience and not just at a distance, not just by observing others and waiting to make the right move. That is probably a way to fall behind. It isn't just experimentation in abstract. Oh, somebody ought to do that. That's not what we're talking about. Everybody ought to be doing that, is my point. Find an opportunity, carve out budget and space to begin testing and utilizing these technologies early and figuring out how your company, how your leadership, how your workforce is going to respond and adapt to those sorts of changes. Kicking the can down the road is not a winning strategy here. At least that's what my team thinks, having studied the history of these sorts of disruptions in the past. Although, again, I'll reiterate, we've never seen anything quite like this before. There's only so much we can learn from looking backward to the past. So those are a few few a few key pieces. Let me mention one more thing that I think is important here. Speaking to the labor market here, it's easy to leap to a very dismal, dark place if you're facing the prospect of losing your job to a machine. Certainly. However, one thing to keep in mind is that this kind of job displacement, that is going to blow the doors off of the production capacity of our economy. And this is very important to keep in mind. So unlike other technology disruptions in the past that have been isolated to just one industry or even one product, like, okay, um, I'm staring up at these LED light bulbs that are shining on my face right now, right? Those have displaced compact fluorescent light bulbs, and compact fluorescent light bulbs in their turn, they disrupted incandescent light bulbs. So those things have changed. But that didn't change all of society and how we structure our lives. It didn't. It wasn't that kind of transformation. Okay. This is something new that we haven't faced before. But through all of that, again, I want to remind everybody to keep this squarely in focus. When human beings cease to do the economically productive tasks in the economy because machines are doing them, all those same tasks are still going to get done. They're just going to get done by machines. What does that mean? It means all the same goods and all the same services will still be being produced. And what that means is that society will still be as wealthy as we ever were before. In fact, machines, at least in principle, should be able to work harder than us and longer hours than us and faster than us. And that means, in principle, they should be able to produce more goods and services than we can with our hands. And so we should be, therefore, wealthier as a whole society in the future, not less, even if fewer of us are working. And what that means is that if we're wealthier, if we're more prosperous, if there are more goods and services, we should be able to find a way for everyone to be better off. Why is that important? Not just because, hey, there's a silver lining on this, not just because, okay, there's cause to be optimistic, but because that means there's a path through this. It means don't panic. We can figure this out. If we're wealthy and prosperous enough, we can work out a good solution for everybody. We can figure out something that works. So don't go to a dark place is my advice with all of this.

Amanda Cupido 23:27

I am a glass half full kind of person too when it comes to this. And I do ultimately think it is going to increase the quality of life around the world. But with that, what important question comes down to purpose. So many people link the purpose in life with getting up in the morning and going to work. So if people have to work less, how are they going to be able to tackle that underpinning of finding purpose in their life in this new era of the earth?

Adam Dorr 23:57

For sure, this is a challenge. Absolutely. I mean, like you said, there is a very large role that our work has played in our identity as individuals. And this has always, you know, this has been the case throughout history, for sure. I mean, heck, many of us have a last name that is an occupation, a profession, a vocation, right? You know, how many people do I know who have the last name Carpenter, right? Uh for just to take one example. So, I mean, it has literally defined our identities. Our names have been defined by the work that we've done. So I don't want to dismiss that as this is a real challenge. And we're going to face a sort of a crisis of meaning individually and collectively, if it turns out we can, even if no matter how prosperous we are, that we're going to have to figure out how do we make our lives meaningful. Okay. So that's on the one hand is yes, this is a real concern. But here's why I'm confident we can figure it out. And that is there have been rich people throughout all of history. There have been lords and ladies, aristocrats, kings and queens and princes and princesses and empresses and all the rest of it throughout history. Not very many of them, okay, a tiny fraction of society in the past, because societies were so much poorer in the past, but there have always been a few people who didn't need to work for a living, who weren't waking up and punching the clock and, you know, and working all day and then punching out at the end of the day. There have always been people who've had that luxury. And they have found ways to have meaningful lives. Now, how have they done that? Well, in a variety of different ways that some of us are familiar with, you know, plenty of folks have hobbies and so forth. But I think the real way that we find that meaning is in each other. The real way we find that is in with our families, with our friends, with our communities, what we give back to one another, what we build, what we create with our relationships with one another. And maybe that's a cheesy answer, and maybe I'm just a little bit romantic. But when I honestly, when I think about, when I think about the the individuals who've led very purposeful lives who haven't had to work for a living, that is what gave them their sense of purpose and their meaning. Another way to think about this is there are two stages in life, even for those of us who aren't, you know, fabulously wealthy and don't have to work for a living, there are two stages in life where we don't work during our childhood and in retirement. And you know what? Childhood can be some of the most meaningful time of life, right? But what do we do in childhood? We learn, we play, we love, we build friendships, right? That's one of the most beautiful things about childhood. And in retirement, ideally, if you're fortunate enough to have good health, for example, then it can be a little bit of a return to something like childhood, where you do more of the sort of things that you wish you had time for: learning, hobbies, playing, and of course, love, the activities with your family and friends and loved ones. So are we headed to a world with it as more like childhood again, or more like retire? Everybody just basically just everybody gets to pretty much retire, or mostly now I'm thinking of my father who's tried to retire three times and then hasn't taken. And he said he's back at it for a fourth time. People struggle, they do. That's a personal example of somebody who's struggled because the life wasn't meaningful enough without having that central role at workplace. But I know we can do it. I'm confident we can do it. And then let me just close with that thought with a fun expression that my was my great-grandmother actually, but my grandmother used to say it too. She used to say, other people would be lucky to have our problems. There's a lot of wisdom in that. Okay. This crisis of meaning, geez, what do we do? What do I do with myself if I have lots of free time and I'm not all stressed about out about where my next meal is going to come from or how to keep a roof over my head? If we're living prosperous and abundant lives because society is very wealthy and a lot and very heavily automated, you know, I'm in sort of forced retirement, but it's a luxurious one, that's a good problem to have compared to other problems that we, you know, certainly that our ancestors had. So I'll take it. I'll take that problem a hundred times out of a hundred.

Amanda Cupido 28:11

Yeah. Yeah. Okay. What about the folks who say as machines and automation take over our tasks, the human cognitive abilities will go down. You stop thinking critically because machines are doing everything. What do you respond to that?

Adam Dorr 28:26

I think this is a real risk. I do. And and I think we're already seeing that in in early data. And nobody has a great answer to this yet. So this is a significant challenge. When we no longer have to challenge ourselves intellectually, cognitively, for a variety of reasons, whether it's because we're not challenging ourselves in a demanding workplace, whether we aren't challenged to solve problems ourselves because artificially intelligent machines are solving problems for us. You know, there is a there's a risk that we could, our minds could atrophy the way that muscles could atrophy if you're not using them. So I think this is a serious, a serious risk. And literally following the atrophy example, we have seen that technologically. You know, our ancestors were physically hardier, more fit than we are, because they couldn't be couch potatoes. They didn't have desk jobs, they were out in the fields. Now, that doesn't mean they were healthier than we are in every respect. I'm not saying that. I mean, they didn't have the benefit of all of our medicine and medical technology and our great food supply and so forth. But, you know, they were using their bodies. Life was too hard to let anything physically go soft. We've seen that happen. We've kind of gone a little physically soft compared to our ancestors. Could that happen to our minds too? Yeah, I think it could. Now, is there any reason to hope that maybe we could there are solutions to that challenge? I think so. For example, let me flip that whole historical example around. There Are more super fit athletic people today than ever there were historically? Our athletes could just smoke the greatest athletes of ancient Athens. And the, I mean, our Olympians would totally obliterate the Olympians of the year 200 BC, right? And, you know, fit people today, they have the benefit of all of our knowledge and all of our technology and all the rest of it. So it is a choice. It's a choice in a way that it wasn't in the past. So the opportunities are going to be there for us to stay even more mentally sharp and be more mentally capable than we are today or were in the past, because there will be better tools and technologies and medicine and health and food and all the rest of that stuff. We will have the opportunity to stay sharper and stay more focused. We will know better how to do that. And, you know, I mean, if we're talking 20 years, 30 years, we start to get into some pretty radical technological capabilities. We're already seeing medicine in the lab, medicines called nootropics that improve memory, that improve cognition. We're going to see more of that. So, in the same way that if you want to be strong today, we know what supplements you should take and how to balance your protein and which macro and micronutrients to consume and how many reps to do in the gym and all that stuff that our ancestors had no idea about. But is everybody gonna make that choice? Maybe not. So uh I am concerned about it. I think there's the opportunity, but our technology isn't gonna do all the work for us. We're gonna have to make those choices. And I think there are things that we're gonna have to do individually, and maybe a society we can make better choices to. Maybe we can find ways to encourage people to make good choices in gentle ways that aren't like Dirconian or tyrannical. We don't want anybody cracking the whip on us, telling us we've got to go out and go run or whatever, or hey, it's time to do an hour worth of mental calculations to keep your brain sharp. No, we don't want to be living like that. But you know, we we we want to find ways that make staying sharp rewarding. And maybe there will be fun new industries, new entertainment, new lifelong educational things, programs, and those sorts of maybe it'll even be new businesses. Find ways to bring that value to people in their lives, both individually and you know, at the family and community level and all the rest of it. And uh it makes choosing to stay sharp a rewarding and valuable choice instead of something like, oh, this is just a chore. So I'm thinking more along the lines of sports leagues that you really want to participate in municipal, you know, municipal data thing, and not, okay, it's time to go out and do some PTs, time to go out and do some, you know, go hit the gym because you've got an obligation to do that. I think we can gamify it, we can make it fun, we can make it part of our lives and our culture to stay sharp and stay healthy and keep those big muscles in your in your skull from atrophying. It's not a muscle, but you know what I mean. Just like we can make good choices about our physical health today. Yeah, it's a concern.

Amanda Cupido 32:49

So much to think about. Wow. Do you feel there's gonna be a tipping point? We talk about this brighter future, but people are afraid of superintelligence, where these machines are just smarter beyond our cognitive comprehension, and this is ultimately what's gonna lead to the demise. What are your thoughts on that? Is there a tipping point that we should be afraid of?

Adam Dorr 33:13

Well, this is a it's a pretty esoteric domain, this issue of ASI, artificial super intelligence, which is different than AI being roughly as capable as human beings are, or perhaps as capable as the brightest human beings are. The real question there is if something is super intelligent, that's like saying, well, what if what if we encountered aliens and they were a thousand times smarter than Einstein? What would we be compared to them? Would we be like, you know, monkeys compared to human beings, or worse, would be we be like chickens compared to human beings or ants compared to human beings? This is sort of sci-fi stuff, but you know, many of the leading figures in the tech space who are leading these tens of billions of dollars large projects, developing the latest cutting-edge AI, they are thinking seriously about that question. And part of it is that there's a scenario in which machines become super intelligent very quickly. The way that scenario works, it's sometimes called an intelligence explosion. This is an idea that goes back a long way to the 1950s or 60s, 60s, 50s. If a machine becomes intelligent enough that it can redesign itself and improve itself, that's sort of a threshold. Like right now, our machines are not really quite able to do that. But once a machine, once an AI is intelligent enough to improve itself, that becomes a flywheel, a feedback loop that can sort of spin up out of exponentially up into the stratosphere, where you have the AI designs an improvement to its brain that makes it even smarter. And then, because it did that, it can design in the next improvement that makes it even smarter, and then even smarter, and so on and so on and so on. And so this is sometimes called a hard takeoff, a, you know, a very fast takeoff, a foom, foom kind of scenario where suddenly these machines become, in some science fiction scenarios, it's like literally overnight. We can imagine, I mean, hey, from the historical perspective, if that happened over the course of five years, in the arc of human history, that would kind of feel like overnight. Yeah, we could find ourselves in a very different kind of situation where we're no longer the biggest, smartest, toughest kids on the block, right? So what does that world look like? And some people are genuinely concerned because those machines, the sort of by definition, would be more capable, more powerful than us and than us, and we would be at their mercy and we wouldn't know what they would want to do in the same way that monkeys don't just can't understand what our priorities and values and goals are. And certainly chickens can't, right? So it's hard to even for us to even imagine like what would these ASIs do? Would they want to keep us around like we like cats and dogs, or would they want to like exterminate us, or would they just not care about us and ignore us? And what would they do? So this is, I don't mean to dismiss these. This it does sound like science fiction stuff, but hey, everything is science fiction until it's not, right? I I just rode in a car that can drive itself out in San Francisco, and it was kind of surreal and it was straight out of science fiction. So we need to take these things seriously now, but I'll tell you why I'm somewhat optimistic, just so I can land on a bit of a high note here and not have it be a eh, you know, that these machines might all be, you know, coming for us and there's nothing we can do. My perspective is overwhelmingly cooperation and constructive activity. So being cooperative and constructive and additive and beneficial, those are all hallmarks, all things that go along with intelligence. So the more intelligent that humans are, in my experience, certainly, and I think at a macro level that you know this is true for all of humanity, the brighter we are, the less we fight, the less conflict we have, the less destructive we are, the less selfish we are, the less short-sighted we are, the less careless we are, all of those things. So the opposite, being thoughtful, conscientious, constructive, collaborating and partnering and pairing and so forth, all of those things are strongly correlated with intelligence. So I'm pretty excited about the possibility of something very, very, very intelligent, being superhumanly, not just good at math or something like that, but also superhumanly thoughtful and considerate and cooperative and constructive and all of those good things. I'm hopeful that that's the case. And then I suppose the last thing that I'll say about superhuman intelligence is that in a very real sense, not figuratively, literally, intelligence is the solution to every problem. You can parse that phrase literally, and it's true. Intelligence is the solution to every problem. We look all through human history, we look forward into the future, any problem we've ever solved, any problem we ever could solve, intelligence is sort of the the key. So if you were to have an intelligence explosion, like I was describing, that means you could have a solution explosion. And that's probably a pretty cool world to be be headed into.

Amanda Cupido 38:18

All right, one final question before we let you go. We'd like to throw a curveball question at our guests. So yours is if robots could dream, what would they dream about?

Adam Dorr 38:30

Ooh, nice. And you know, there was a there's a very famous science fiction story called Do Androids Dream of Electric Sheep? It's a strange name, but it was the it was the short story that the famous movie Blade Runner was based on. So yes, what if if robots could dream? You know, there's a theory, we don't know exactly what dreams are for. They're still mysterious, and it's part of what makes them beautiful. But one of the things that they might be for is to help us learn and remember, and maybe more importantly, take all of the things that we've learned during each day and digest and synthesize and and make them part of ourselves, forge new connections between things that we didn't realize were connected before. So dreaming probably has some beautiful purpose, and biology is very elegant. We've seen our cats and dogs dream when they're sleeping, and so there must be a really good reason for it. And I bet I my guess is that we will realize that dreams are doing something so important for our minds, that evolution has built that into not just humans, but lots and all mammals. We may realize that anything that's intelligent needs to do something like dreaming. And so, what will robot dreams be like and what will they dream of? Maybe they will dream of the kind of future where things today that seem impossible to achieve, problems that seem impossible to solve, can be solved. And so maybe that's a little bit hokey, a little bit cheesy, but that's on brand for me. I don't mind wearing that uh badge. I suspect robots will will probably, and by that I mean artificial intelligence will probably dream or do something very much like it. We will realize that they have to, in order to be intelligent in all of the ways that human beings are intelligent. And we will find that they are probably sources of inspiration and sources of insight and creativity for artificial intelligence, the same way that they're sources of inspiration and creativity and insight for biological intelligence. I guess what I'm saying here is I bet intelligence is what's fundamental. And it doesn't matter so much whether it's biological and your brain is where it comes from, or whether it's artificial and it's silicon chips and binary code that where it comes. It may be that intelligence is what is really what really matters, and then it unifies us. And I bet if dreams are part of what it means to be an intelligent creature, maybe that applies to all of us, whether we're made out of organic molecules or silicon or whatever. So I bet AIs will dream, and I bet they'll dream for similar reasons that we do.

Amanda Cupido 41:28

That is fascinating. Well, thanks, Adam, for all of your brilliant insights, and thank you for listening. This has been People B4 Machines Conversations on the Chaos of Factory Automation Powered by Eclipse. We're here to challenge the status quo in factory automation because machines don't build factories. People do. The technical producer for this podcast is Ryan Dentinger. I'm Amanda Cupido. If you got something out of today's episode, we'd appreciate it if you share it with a teammate, a plant leader, or anyone who's tired of the automation echo chamber. Be sure to follow for real talk, bold questions, and sharp insights. And remember, the future isn't fully automated, it's people powered. Talk to you soon.

Intro 42:12

Thanks for listening to People B4 Machines. Conversations on the chaos of factory automation from Monday morning to the very near future. For past podcast episodes, search people before machines on Spotify, Apple Podcasts, Amazon Music, YouTube, or ww.PeopleB4Machines.com.