Why the energy transition is a ‘delusional’ goal: Policy expert Mark Mills on the myths and realities of our energy technology

Dec 19, 2023

Wind turbines turn behind a solar farm in Rapshagen, Germany, Thursday, Oct. 28, 2021. Michael Sohn/AP Photo.

This episode features host Sean Speer in conversation with Manhattan Institute senior fellow and energy policy expert Mark Mills about his book, The Cloud Revolution: How the Convergence of New Technologies Will Unleash the Next Economic Boom and A Roaring 2020s.

They discuss technology and innovation, the limits of an energy transition, and why he thinks we are on the cusp of the "roaring 2020s."

You can listen to this episode of Hub Dialogues on Acast, Amazon, Apple, Google, Spotify, or YouTube. The episodes are generously supported by The Ira Gluskin And Maxine Granovsky Gluskin Charitable Foundation.

SEAN SPEER: Welcome to Hub Dialogues. I’m your host, Sean Speer, editor-at-large at The Hub. I’m honoured to be joined today by Mark Mills, a senior fellow at the Manhattan Institute and a faculty fellow at Northwestern University's McCormick School of Engineering and Applied Science, who's written extensively about innovation, energy, and climate policy. He himself has a physics background and has worked in the White House science office in the Reagan administration, as well as Canada's own version of Bell Labs.

I first encountered Mark's work a few years ago when I attended a talk he delivered on his book, Digital Cathedrals, about the extraordinary energy demands of modern information infrastructure such as data centres. I’m grateful to speak with him about a range of topics, including recent news of progress on nuclear fusion, his views on the energy transition, and ultimately why he's optimistic about what he predicts will be the "roaring 2020s." Mark, thanks for joining us at Hub Dialogues.

MARK MILLS: Thanks, Sean. Of course, the 2020s are mewing and not roaring at the moment, but they’ll get there. They’ll get there. [laughter]

SEAN SPEER: Let's start with nuclear fusion. We’re speaking the week of December 12th in which we’ve learned of new progress on nuclear fusion out of a laboratory run by the U.S. Department of Energy. You’re a bit skeptical though. You wrote in an op-ed in the New York Post that, "Fusion is always 50 years away." What did you mean and why should we be careful in the way that we think and talk about the progress here?

MARK MILLS: Well, now, it might really be 50 years away instead of always 50 years away. That's what the progress really means. I think in my piece, as you know, I was properly enthusiastic as a former practicing physicist. Once you’re a physicist, I guess you’re always one. It imprints your brain in an indelible yet perverted way of how to think about the world anyway.

It's a really big step in the physics of fusion. People should understand that the facility that did this, it's a 500-foot-long, seven-story-high laser machine. This is a big machine. It's a science machine that costs almost $4 billion to build. It wasn't designed to make commercial nuclear fusion. In fact, it was designed to study the physics of fusion in nuclear bombs just to put an unpleasant spin on what they’re really doing there.

The physics of fusion, how the sun and the stars make energy, and how they make the elements that we use, that's where they came from. Stars that we didn't create created the elements that we use for everything. It's really exciting. They got what's called "first ignition," ignition meaning that the energy that they’ve put directly into the little fuel pellet—which is a tiny, little, eraser-sized area of a derivative of hydrogen called tritium and deuterium. These are basically hydrogen, right? So-called isotopes of hydrogen, or forms of hydrogen.

They got more energy out than they put in by 20 percent. That's a big deal. It's called ignition. It's like when you start your car, you have to have an ignition. Okay, you have a spark plug, it gives you ignition, assuming you have the kind of car that 99 percent of people do, but anyway. [chuckles] But then you still need a car. You still need the rest of the system. Of course, they only got ignition, so that's a big deal in physics.

But the leap is from there to building a power machine, which is why I was saying it's 50 years away. To give you a sense of why it's a long way away, the lasers that they used to do that consumed 200 units of energy to produce the one unit of fusion energy. The grid power was still 200-fold more than the output. To make a proper energy machine, you need roughly speaking 100 times the output of what your input was.

Otherwise, you’re not producing energy. You’re consuming it, right? That's problem one. Those lasers don't exist. We don't know how to build them yet. We’ll figure it out. Engineers are really good at this kind of stuff. We’ll figure it out. Might take 10 or 20 years. Another 10 or 20 to build them at scale. Then the fuel pellet, this is important, the fuel pellets are made like jewelry right now. They’re handmade. They cost about $1 million each.

That machine uses up a few hundred fuel pellets a year, so you could do the math. Maybe 500 or something like that. You’re going to need about 10 fuel pellets per second to make a power machine, not a science machine. Those pellets, obviously, they’ll have to cost a lot less than $1 million each if we need millions of them a year per machine. That's an engineering problem. That's not trivial to manufacture things like that at those kinds of scales. Again, we’ll figure it out.

The two buckets that I was trying to get people's heads around were the science bucket. It is a big deal to conquer these science barriers. Then the engineering bucket to make machines that society can use. It takes a while. I wrote in a piece, and it's in my book as you know, that from the first internal combustion engine to the first practical car was about 50 years. The first steam engine to the first practical trains was about 50 years. From the first photovoltaic cell to practical photovoltaic cells, pretty close to 50 years.

From the first conventional lithium chemistry for a battery to practical electric cars, about 50 years. It was the mid-’70s when the lithium chemistry was invented. This is a timeline that has not accelerated for 200 years. This idea that we have accelerating technology, that's only true in the domains of information technology. Energy technology, big machines to run society, they operate on time scales at civilization levels that are measured in many, many decades. That's just the way it is.

SEAN SPEER: We’ll follow up with some of those observations when I come to you specifically about the energy transition. Before we get there, though, Mark, your op-ed sets out three fallacies that can influence how we think about breakthrough technologies, including with respect to energy. What are they and how do they distort our thinking?

MARK MILLS: Well, as I look at what people say and the questions I get, what policymakers write and what pundits and prognosticators are excitedly babbling about in the chattering classes, you look at all this stuff and you try to see patterns in it. Well, this is another fault of being a physicist, but psychologists do the same things. It's not just physicists. You look for patterns. We all do. Humans are pattern-seeking beasts.

Patterns have meaning, so I distill the pattern of fallacies and understanding energy technology into three rules. The first one that I did was the magic wand fallacy: that we could come up with a new technology—again, this is about energy; this is true with lots of technologies, but energy in particular—that one new technology is going to solve all our problems. No, we still burn wood.

In fact, the world gets 300 percent more energy from burning wood today than all the wind and solar in the world combined. We, in fact, burn more wood now than we did 100 years ago to make energy. Wood has not gone away, so the magic wand of coal and oil then gas didn't eliminate the use of wood, in fact, even. There's no magic wand that you can solve all the problems or make things happen overnight. They take a long time as I just said.

The second fallacy I called the helicopter fallacy because I happen to like machines. I think helicopters are poorly appreciated as one of the most amazing machines humans have invented. In fact, there's a great book for those who like the history of technology called The God Machine. It's about the invention of the helicopter because making machines that can fly humans required wings to basically emulate nature, right?

The idea of a helicopter dates back to some drawings that Leonardo da Vinci had, but no one is ever going to make a heavy machine that could just take off vertically. That's a helicopter. Really amazing. When commercial, practical helicopters were invented, the chattering classes were just all aghast and babbled and, "It's going to change everything. It's going to eliminate cars and airplanes. Everything's going to be different."

Well, apparently not. It turns out they have a very useful niche. It's a multi-billion dollar industry, but you would no more use, as I wrote, a helicopter to fly across the Atlantic than you would use a nuclear reactor to power a train, or, to be facetious but true, use solar panels to power a country. Some things just don't make sense, and that's the helicopter fallacy. They have utility, but it's specific.

The third fallacy, it's an easier one to describe. It's the moonshot fallacy. This is true of lots of stuff from cancer to climate change. Every politician loves to call it the moonshot or they pervert it and call it the earth shot. The moonshot. Well, it was really impressive. I’m a moon junkie. I wanted to be an astrophysicist when I started my education at Queens University in the Great White North of Canada. [chuckles] By white, I mean snowy. I’m not making a race statement here. It's a snowy country.

Anyway, I love this stuff, but putting a dozen men on the moon—and they were all men then; there’ll be women in the future—putting 12 men on the moon once doesn't change everything. It's an amazing feat. The analogy in energy would be we have to do the equivalent to put all of humanity on the moon. We know that's not possible. We know how to put dozens of people on the moon. We might even put a hundred people on the moon, maybe a thousand, but all of humanity? Not going to happen, except in science fiction. That fallacy is really important in terms of what technology can accomplish and what engineers can do, especially in energy.

SEAN SPEER: That's a good segue to my next question, Mark. Let me set it up. You’ve written extensively about this so-called "energy transition" and national and international goals of net zero emissions. You wrote in an August paper for the Manhattan Institute, for instance, that "the energy transition is not feasible in any meaningful time frame and it's a dangerous delusion to base policies on the idea that such a transition is possible." Let me start with a two-part question. First, why is it infeasible? Second, why do you think so many governments have come to base their policymaking on this infeasible idea?

MARK MILLS: Well, the second one is the political and psychology question, not the physics question, but I’ll take a stab at it in a second. The first one is, why isn't it going to happen? That's why I’ve written a lot about it, why you have to keep writing about it, because people keep saying things that are just—again, I’ll do the rule of threes. I like threes as from my recent book.

There are three buckets for technology and energy. There are some things that are just impossible that just won't happen. You have to understand the physics and know that some things are impossible. There are some things that are infeasible, not impossible, but we really don't know how to do them. Then there are some things that are just silly. We can do it, but we shouldn't.

Those are the three categories—other than the things you ought to do. But the three categories of things that won't happen are those. When I look at the claim that we’re going to eliminate hydrocarbons, we are going to reduce the use of hydrocarbons as a percentage of the world's energy. We have been. We spent $5 trillion in the Western world in the last 20 years to avoid hydrocarbons.

We have reduced the percentage of the energy the world gets from hydrocarbons. It used to be 84 percent 20 years ago. Now, it's 82 percent. It's a pretty good reduction, $5 trillion, two percentage points later, but it was a declining percentage. It wasn't a decline in absolute use. The absolute quantity of hydrocarbons has increased in the last 20 years by an amount equal to adding six Saudi Arabia's worth of oil demand to the world, so it's a pretty big increase.

That alone should tell you something, right? If you look at the underlying engineering requirements, the material requirements, the cost requirements, the hug-and-get-along-that-we-all-have-to-do-the-same-thing requirements on the whole planet. The constellation of requirements would tell you, it's not going to happen in any kind of time frame that has any meaning. You have to dissect what's in their head that they have tropes about, "Oh, it’ll go faster. They’ll get better. Technology can do this, can do that."

In every case, you have to dissect it based on the three metrics, right? Can't happen ever. Batteries don't get better as good as oil and energy density. Just won't happen in the physics of the universe we live in, period. Batteries work just fine, but they’re a really expensive way to store energy. Using a lot of them for all energy is just infeasible. It's not going to happen. Pushing them as an alternative to regular cars is just silly because it's expensive, not that you can't do it. All these things fall into those three buckets.

I write about that. People complain that you’re not optimistic or you’re not creative or you don't recognize all the progress. Well, no, I do. I studied this obsessively, more than I should, but people are being either naive or insufficiently curious about the underlying facts, or silly. There are lots of problems, which brings me to answer the question of the psychology and why do governments do what they do?

Well, governments do a lot of really silly things. Both the Canadian Parliament and the American Congress, and thankfully because we’re democracies, have the right to pass laws that just violate the law of nature. They can violate the laws of common sense. They can violate the laws of economics. They do it all the time, but I would rather live in a system that's democratic where we try things that are silly, then we pay the price and reset, rather than live in a dictatorship where the mistakes—and you can guarantee their mistakes because the dictators are no smarter than the democratically-elected people. In fact, maybe they’re dumber, but let's just assume they’re equally smart. Everybody's going to be wrong about the future. You want to have the flexibility to adapt around what's actually possible. That's why I write in the policy space.

I don't like when things are done that are really, I think, potentially immoral and damaging. Europe, the reason it's in the fix it's in, it can be reasonably laid at the feet of not Putin. Putin is an exogenous variable. The fact that Europe is so unable to respond to the loss of energy from Russia is because they’ve put so much of their money into non-hydrocarbons. They don't have surge capabilities. They don't have flexibility. They’ve made a fragile energy system by depending too much on wind and solar.

That fragility is costing them both politically, economically, and socially. Building more wind and solar won't reduce that fragility, the economic and geopolitical fragility, it will increase it. I think politicians in Europe now know that. A lot of them do. I think they really do. I don't think they’re admitting it so much publicly, but I think they are privately facing up to those very harsh facts, which are very consequential, which is why I say it's a dangerous delusion because the consequences are dangerous.

They’re dangerous to human life, they’re dangerous to our geopolitical status, and they’re dangerous to our economies if we make energy decisions that don't work because it's beyond obvious to state that nothing exists, nothing is possible without energy. Everything requires energy. Because of that, you want energy to be two things: reliable and cheap. Those are utterly critical and sacrosanct metrics for a healthy society. The energy path that Europe put itself on made energy more expensive and less reliable in every sense.

SEAN SPEER: In an interview with Bill Kristol in April 2022, you talked about the magnitude of investment in—

Mark: Looks like you’re like a stalker.

MARK MILLS: Nobody watches Bill Kristol. No I’m kidding.

SEAN SPEER: Let me set this up, Mark, because the analogy that you drew there has really stuck with me in the sense that it's helped to convey the magnitude of investment and progress that would need to happen to achieve an energy transition by 2050. I’m mostly paraphrasing, but you said something like it would involve a 9,000 percent increase in the infrastructure of wind, solar, and batteries in the next 20 years or so.

To put that in some context, you explained the state of wind, solar, and battery infrastructure today is about the equivalent of where it was for oil and gas in 1950, which has grown by 1,000 percent over the past 50 years or so. In other words, a 9,000 percent increase in 20 years versus a 1,000 percent increase in 50 years.

As I say, Mark, that analogy has been something that has stayed with me when I think and read about some of these questions. Why don't I just ask you to elaborate a bit on your insight here and, more generally, the impracticality of so-called "getting off oil and gas"?

MARK MILLS: The reason I did use that analogy is to get people's heads out of thinking that there's something magically different about wind, solar, and batteries, or hydrogen. Pick your poison. It doesn't matter, because everything requires building physical infrastructure. All energy machines have to be built by mining something, making steel, concrete, glass, or getting copper, molybdenum, zinc, rhenium.

All kinds of metals and elements are needed to build these machines. It takes time. It takes backhoes and Caterpillar D9s and John Deere equipment and takes trucks and people and cranes. It's all the same. Let's just even take inventory approval out. Just assume that we could do things as easily as we did in the 1950s because the regulatory footprint was lighter then. It was obviously easier to build things fast. We built things then the way the Chinese do now basically.

Look at the Empire State Building, infamously built, or famously built, depending on your view of skyscrapers, in less than a year, right? Could we do that today? Well, Germany just built a skyscraper-class piece of infrastructure called a liquid natural gas import terminal in eight months, not eight years. Eight months, they did it. They went from starting it because of the invasion in Ukraine in April and they just opened it up last week.

We can still do stuff like that but set that aside. That's a whole separate point. Assuming we could do stuff like that still as we did in the ’50s, the physical stuff that we build to make energy, the hydrocarbon energy, in total energy supply terms in the 1950s, to your point, was equal to the physical infrastructure that we have built today globally for wind and solar. The amount of energy supply to the world in absolute terms, not in relative terms.

The wind and solar industry is very big today. It's only a few percent of the world's energy, but that's very big. What I looked at was a very simple construction question. How much did the oil and gas infrastructure of the world grow, the physical stuff, from the mid-’50s over the next 50 years? It grew by 1,000 percent. It's a lot, right? The world's demand for the hydrocarbons led to a construction program that increased the infrastructure by 1,000 percent in a half a century.

We’re starting from the same point, doing the same thing. We’re building stuff. It's the same point as oil and gas in the 1950s. We wanted to expand, and it will. Let's just say we wanted to expand faster, to your point. Well, to get to the goals people have in their heads, we have to expand it not by 1,000 percent but by 9,000 percent from the same starting point, same kind of hardware, same kind of trucks. We want to do it not in 50 years but in 20 years.

Okay, there's no evidence that the construction industry can move that much faster. What I didn't say, I don't think, in the interview with Bill Kristol was that to get to the same amount of energy increase, it's not just 1,000 percent versus a 9,000 percent increase in infrastructure. It's actually even bigger than that, because to produce the same amount of energy from wind and solar, the physical infrastructure, the hardware, and all the stuff, roughly speaking is 300 percent bigger.

Do your elementary school arithmetic. That would mean we want to do in a construction sense something like a 10,000 percent increase in 20, 25 years of the same kind of activity as we managed in 50 years to do 1,000 percent. Cut the time in half and increase the quantity of stuff we’re building like tenfold. Okay, I don't know. I’ll take the bet. Not going to happen. It's just not going to happen. That's why I say that's not going to happen.

"How could you be so sure?" Well, show me. Show me the construction program, the machinery, the way we can make concrete and move steel and build wind turbines the size of the Eiffel Tower faster now just in terms of tons of stuff per day than 50 years ago. We’re a little faster and we’re going to get a little faster because we’ve got really good machines, but we’re not 1,000 percent faster. We’re just not. It's silly. That's why I use the provocative word "delusion" for these paths. They’re delusional. Even though they look good in PowerPoint, they don't exist in the physical world.

SEAN SPEER: Now, Mark, I suspect that most of our listeners will be open to and persuaded by those arguments. The facts, of course, don't lie. Some may respond negatively however to your argument that these alternative forms of energy can only be a supplement rather than a replacement and what we actually need to do is boost oil and gas production. What's your argument? Are you saying that we need to, in effect, trade-off our goals of emissions reductions for economic and geopolitical reasons, or do you think we can still make climate progress even as oil and gas production grows?

MARK MILLS: First of all, I don't speak in terms of the language of climate progress for a whole lot of reasons because it takes us down a rabbit hole of different issues. My focus is not about climate progress per se but pointing out that, for whatever reason, it doesn't matter and it makes no difference what you think about carbon dioxide emission's effect in the next 50 or 100 years or today, if you think it's having an effect today, which is disputable. Everybody thinks it's having an effect today.

It makes no difference. It doesn't affect the physics of energy. It doesn't affect how fast a D9 can move or how we make concrete. It has no bearing whatsoever. The only relevance is one direction. If you’re saying, "I don't want to burn hydrocarbons," you have to tell me what you want to do instead. Then you have to be honest about what doing that means in dollars, in geopolitics, in environmental impacts and water use, in real pollution, and who we buy from.

What I’m trying to do is illuminate the trade we’re making. If you don't want to use the liquids and gases that we use today for 80 percent—well coal gets you to 82 percent, so take coal out. Oil and gas are more than half. If you don't want to do that, you have to be honest about what the real-world consequences are of the alternative, and then you have to ask yourself if you can actually do it. That's why I’m doing that.

The different question that gets asked—I’ll rephrase your question this way, because if the issue is not climate progress, the question really is, all right, if you are honest and admit there is nothing you can do to reduce the emissions of carbon dioxide—you can slow the rate of increase through depression, recession, and deprivation. That's one way. We learned that with lockdowns. You can actually just very slightly reduce global energy use. It was just very slight.

It's amazing. Shut the world down and you get a few percentage point reduction in world energy use. You get a lot of gasoline use going down, but world energy use didn't go down very much. You can do that. There are people who really want the deprivation path. They say, "We have to have de-growth now." Let's be honest about that, or you’re going to have to admit that carbon dioxide emissions are not going to go down.

What they’re going to do is what they’ve already been doing for decades. They’ve been doing it in the two decades of so-called climate awareness. They’re going up. Then the question is, "what do I do?" My answer is not facetious. It's honest. We have to be patient because anything we want to do to change energy infrastructures will take far longer than policymakers and pundits are proposing. It's just going to take time.

If we’re going to take time—that's both for the new science, better technology, all of it—it will take time changing infrastructures. If it will take time and if you in your heart of hearts believe there will be negative consequences, then you have to be honest about resilience. We should be spending our money. The second-most precious resource in civilization after our time is our money. They’re intimately-related phenomena, which is we should be spending our time and money on resilience.

We should be protecting infrastructures and people and societies from the consequences of nature's deprivations even if we didn't cause them. If we did cause them, then all the more reason to protect people. We can't afford to squander money on things that don't help when we should be spending money on things that protect people. A lot of the climate apocalyptics reject that as giving up.

It's not giving up. It's actually acknowledging that you think there's a consequence and protecting people from the consequence by spending money on that kind of engineering. It's not a satisfactory answer, but it genuinely is an independent magisterium. The climate science debate is only relevant to the extent that it's motivating policies that are damaging, that are not possible. That's the relevance. The relevance is not the inverse.

What do I do to make climate progress? It sounds facetious to say you can't. If you’re honest about what the data show, you can't. If the United States and Canada ceased to exist tomorrow and emitted zero, we imported nothing, we just evaporate, and we didn't import carbon-intensive products, we emitted zero, we all died and went away, the world's carbon dioxide emissions are going to go up. Because of what the rest of the world is doing, they aren't following our path and they can't.

This is the point is they can't follow our path to zero. We aren't on a path to zero. The CO2 emissions reductions that have occurred in some countries are actually elusive and not real in the sense that—this is why Europe is turning to a border adjustment tax—it's because what we’ve done in North America and in Europe is outsourced the energy-intensive and, therefore, carbon-intensive industries, especially mining and metals industries, to parts of the world that are willing to do it, and they use hydrocarbons there.

SEAN SPEER: Let's shift the conversation, Mark, in our remaining time together to innovation, technology, and economic growth more generally. In the face of growing commentary about decadence and stagnation, your most recent book, The Cloud Revolution, sets out a far more optimistic story. You write about the convergence of different technologies that will, in your judgment, produce something like the roaring 2020s. What's your key insight that the stagnationists are missing?

MARK MILLS: It's a tough market to be an optimist in right now. [laughs] I stipulate two things. First, the book's time horizon is the next decade or so, deliberately, not the next 50 to 100 years, not the next 10 months or election cycles. Of course, I wrote in my introduction to my book and I ended my book with this in the epilogue. Governments can Sovietize economies. They can destroy economies even during periods of potential growth.

Politics matters a lot. This does. Getting the politics right matters enormously. I think we have a decent shot at getting it right. We just often get it wrong for a while. The reason I’m optimistic is I would say, to put it into simplest terms, technology doesn't make progress in linear, even steps. It happens in spurts or waves. Waves aren't cycles. People have in their head cycles, they’re predictable in the sense every 10 years or every 50 years.

That's not what happens. We have waves of innovation. If you look over history, the Industrial Revolution was in one sense a wave. And then the second Industrial Revolution. There was an Industrial Revolution in the Middle Ages. That was a wave of innovation around the windmill, at the time, and the watermill, and the development of a camshaft and the gears and pulleys. We’ve had periods of innovation.

That's the nature of the beast. That's just how life is. It doesn't occur uniformly and it occurs in spurts. We haven't had a wave of innovation as foundational as it occurred a century ago. Basically, that matured in the 1920s. At that time, that was the time, not the invention of but the maturation, of innovations in three different distinct domains. They all happened simultaneously.

In the machine domain, we had the car and the airplane and the power plant to pick three obvious ones. None of which were invented in the 1920s, but all of which matured then. This is true of all technologies. When they matter is when they’re maturing, not when they’re invented. When they begin to mature. Then if you think of the materials domain in the 1920s, we had then the maturation of high-strength steels, high-strength concrete, polymers, and pharmaceuticals.

These were materials revolutions. We started making things out of stuff that we’d never made things out of before. We made things that were artificial at scale. Artificial fertilizer, which revolutionized food production, all happened in the 1920s. All invented in the decades before but matured then. Similarly, in the 1920s, we had the professionalization of science that occurred for the first time.

We had the information revolution around science, but the information revolution, or the maturation of telephony, that's when it took off. Radio took off then and TV began shortly after. These things all happened. They’re all independent. If you think about every one of the things I talk about, historians will write about the history through the lens of each one of those. What's interesting to me is all of those happened contemporaneously.

That confluence is what gave us the explosive economic growth of the 20th century. We still had wars. We did bad things. We had the Soviet Union. All sorts of grotesque things happened. Underlying that was the biggest expansion of human wealth in all of history. The percentage of human beings that are in abject poverty collapsed in the 20th century. There are still people in abject poverty, but the percentage in abject poverty collapsed.

If you use that metric and look at the same three domains today, what I’m proposing—and I outline in my book in detail because it's easy to tell the history of the way I did, the future takes more time because people don't believe you until you give them the taxonomy. If you think about it, think about it in those three domains of things that were invented in the last decade or two but are just now maturing and haven't quite taken off yet.

In the machine domains, the two things that are the most obvious are the capacity to manufacture things at a molecular atomic scale, that we actually build things from atoms up, which is what 3D printing is and molecular machines are. The transistors are basically molecular machines. We build molecules and atoms. This is an unprecedented capability. It's just beginning to bleed into every part of manufacturing. Just beginning to because we just figured out how to do it in the last couple of decades at scale.

Then, of course, autonomy is the biggest machine difference. Robots in a word, but real robots. A washing machine is a robot because it's autonomous. You don't have to stand there. You put stuff in, you leave it. When you say robot, everybody in their head knows what I mean. I want something walking around and helping me, okay? Is that kind of robot possible now? We know it is, because you can lease one in the business world. You can lease walking robots. They exist.

Now, how many of them are in the world? Only a million, but there were zero 20 years ago. There are a million of them now. There will be hundreds of millions. They’re profoundly impactful. Drones would be another category of autonomous machines that are just beginning to show up. The machine world is just on the cusp of changing. The materials world is changing. The words themselves tell you a lot about them.

They’re self-healing materials. They’re self-assembling materials. We make biocompatible electronics, which means you can swallow things that would resemble microscopic computers to do diagnostics in real time. These aren't imaginary things, they’re things that exist. Of course, in the information world, the glue of my book, which is why it's called The Cloud Revolution, is on the information side.

We’ve done something that hasn't been done in a long time. We built a new infrastructure. The cloud is not a computer infrastructure. This is the conceptual link that's needed. Yes, it uses computers, but computers compute. The cloud provides advice and inference. When you ask Google Maps to do something, it gives you advice. That's a computer route. It gives you several routes based on traffic and, increasingly, weather.

This is true in manufacturing. Now, we have natural language computing that's resonant in the cloud that's accessible anywhere that gives advice. Advice-giving is really, really different and consequential. It's a powerful, powerful amplifier for human capacity and capabilities, and it's really disruptive. I’m not naive. Technologies all have negative effects. AI will kill jobs like the automated loom killed the jobs for the Luddites.

Of course, it did. Of course, drones can be used for weapons. People will always fight. They seem to want to. The constellation of those three revolutions are going on now and that they’re all confluent simultaneously is really unusual and extremely powerful and will lead to a wealth expansion in the rest of the 21st century that will be equal to, I think, or greater than what happened in the 20th century.

The epicenter of this is North America by the way. I say U.S. in my book because—I’m a Canadian and all that—but it's U.S.-centric. Waterloo is incredible. I know the Blackberry came from Canada. Believe me. I remember. I was at Bell Labs in Canada. We put the first communications cable in with the fiber optics in Canada. Canada was the home of the first radio. Canada has a role here.

When I say U.S., I mean the North American ecosystem is the epicenter of all this innovation. Not the only place, but it is the epicenter. It's very bullish for America, not for China. Our demographics are better. They’re not perfect, but they’re a whole lot better than China's. They’re better than Europe's. Europe's just disappearing into itself. Fundamentally, I’m bullish on North America. I’m bullish on the economy long-run.

I’m bearish on China, but what happens in the next year or two and we—before my book was published, I wrote the obvious: there’ll be another war. Sure enough two months after it came out, Ukraine gets invaded. People do that. They do bad stuff. We got bad people in the world. My underlying theme in the book is to illuminate the structural shifts that are going on across all these domains and not based on idealistic, "Oh, it would be nice if we had X," or "Shouldn't we have Y?" or "Shouldn't we have a cure for cancer?"

I don't predict the cure for cancer by any means. I predict the fact that we’ll be much more likely to find it now because of artificial intelligence and AI and the kind of tools we have. When I say we can make materials at the molecular scale, there are commercially viable means to fabricate so-called nanomachines. It's machines that are smaller than cells that are harmless to you that you could put in your bloodstream that will physically drill a hole in a cancer cell.

In other words, instead of bombarding your body with chemicals, the machine identifies a cancer cell, lands on it like a drilling machine, and drills a hole in the wall of the cancer cell. It injects a lethal dose into the heart of the cancer cell instead of all the healthy cells and then evaporates. It self-destructs itself because it is programmed to disappear. Those kinds of science fiction things are found in things that exist today.

The whole theme of my book is around forecasting based on an aphorism that I steal from the management consultant, Peter Drucker. I have an appendix on forecasting, but Peter Drucker said he stopped predicting the future when he made a prediction in 1929. The stock market was going to go up. He made the prediction on, literally, the eve of the crash, the biggest stock crash in history. He said he spent the rest of his life only predicting things that had already happened.

It was a cute line, but it's a meaningful line in technology. Because if you look at the pantheon of things that have already been invented, not in a lab bench but moved into commercialization timelines, that they’re either about to become commercial, like a computer you can swallow to do blood diagnostics or air taxis. Well, there are not going to be an air taxi for a while, but we have drones that work just fine and carry cargo. There are going to be thousands of them deployed in Africa.

They’re already deployed in Africa to deliver medications and high-value supplies. What's already been invented and is just beginning to commercialize or is about to be commercialized tells you a lot about the near future much more than a patent or an aspiration. I’m entirely focused on the future of what's just been invented across those three domains I described.

SEAN SPEER: Mark, I know you have to go for another interview, but this has just been a fascinating conversation as is your book, The Cloud Revolution: How the Convergence of New Technologies Will Unleash the Next Economic Boom and A Roaring 2020s. I have so many other questions to put to you. We’ll have to have you back on the program. Mark Mills, thank you so much for joining us at Hub Dialogues.

MARK MILLS: Thanks, Sean. Happy to come back anytime and let's keep our countrymen warm in Canada.