October 6, 2022
How does 3D-printed food work? How do hackers and inventors think? What are some ideas that don't matter? Why are humans so driven by stories? What are the current sentiments around nuclear energy? What is an "information DMZ"? Is "cryptocurrency regulation" a contradiction in terms? What are "deep" and "shallow" technologies? How could we handle intellectual property rights more fairly?
Pablos is a hacker and inventor that runs Deep Future, a venture capital firm backing mad scientists, rogue inventors, crazy hackers, and maverick entrepreneurs who are implementing science fiction, solving big problems, and helping our species become better ancestors. Pablos is a top public speaker on technology whose TED Talks have over 30 million views. With his Deep Future Podcast, Pablos is sharing his conversations with people who understand the biggest problems in the world and the technologies that could help us solve them. Follow him on Twitter at @pablos, email him at email@example.com, or find out more about him at deepfuture.tech.
JOSH: Hello, and welcome to Clearer Thinking with Spencer Greenberg, the podcast about ideas that matter. I'm Josh Castle, the producer of the podcast, and I'm so glad you've joined us today. In this episode, Spencer speaks with Pablos Holman about invention and hacking, the importance of stories, and deep technology.
SPENCER: Pablos, welcome.
PABLOS: Thank you.
SPENCER: Yeah, I'm really glad you come on. Since you're a person of many ideas, I think the way to approach this is just to go bang one idea after another and see how many we get to. How does that sound?
PABLOS: Sure. We can get through a lot.
SPENCER: Yeah, I think we will. So the first topic for you: you came up with this idea of 3D printing food, tell us about that.
PABLOS: Well, I had been working at a lab called Intellectual Ventures, which was a lab to help build for Nathan Myhrvold. What we were doing was working on a wide variety of different kinds of invention projects, just trying to invent new technologies. And so this lab is a totally unusual place. Unlike any other lab, we had been working on inventions, which is different from basic research or scientific discovery. So, we had electronics labs, laser labs, food labs, bio labs, and chem labs (more like one of everything), and then one of every kind of scientist. And then we had a prototyping machine shop. We basically just bought one of every tool in the world, and so we could make anything in the lab. But what was interesting is one of the major projects going on was a team called Modernist Cuisine. They were doing work on the science of cooking, and they're famous for having built a huge publishing about their cooking techniques and figuring out new techniques for chefs and things that people could do with the kitchen. So the Modernist Cuisine team is known for being like the biggest heretics in food because they're advancing scientific techniques and equipment to try and help chefs understand at the molecular level what's going on, and then give them new capabilities. That's usually called “molecular gastronomy” or “modernist cooking.” And so I had a team of food scientists and chefs and chemists there. And then right next to them, we had a machine shop with all kinds of machines that could build things, including 3D printers. — And this is over a decade ago, so this is at a time when nobody even heard of a 3D printer. But I had 3D printers and a machine shop, so I started trying to co-op people from the kitchen, and said, “What if we do post-modernist cuisine, and we take these 3D printers and make them print food?” It sounded crazy even to these folks who are, in some sense, like the craziest people in the world around cooking. And Nathan Myhrvold is well known for being a heretic, in science in general, but in food. — And so we started coming up with a bunch of ideas for how we could make robots make meals or make 3D printers print food. And the idea there is very important for two reasons. One is the way we feed humans right now is just wildly inefficient. You basically just grab some raw material and blast a bunch of heat at it, and then eat it. And that's not very efficient because you throw out about 40% of the food on average. And in different parts of the world, you throw it away at different stages, but Americans throw it away after we buy it and cook it and serve it and then can't eat it all and put it in the fridge for two weeks, and then throw it away. So that's our version of inefficiency. So you could solve all that. Because imagine if you had a machine that had ingredients in it, that was wonderful. Imagine you grow a tomato on the vine, but it's genetically selected for flavor and nutrition. And it's this beautiful, juicy red tomato and then you pick it, flash freeze it, powder it, put it in a sealed toner cartridge, and send that to your 3D printer, which then prints a pixel of the world's best tomato, hydrate it with a needle, zap it with a laser to cook it, rinse and repeat for every pixel and now you're printing this meal with the best ingredients in the world, but you're not shipping any water around the planet. The ingredients are shelf-stable for years, they have all the flavor and all the nutrition, so now you're not wasting anything. It only prints as much as you're gonna eat. And so you eliminate the waste. But the other thing that's awesome about this is that you could make meals that are customized for people, where for the first time, you get the nutrition that's optimized for you. And we're getting so much better every day now at being able to understand your microbiome and what's going on in your gut and be able to design an optimized diet for you. We don't have any way of making customized meals. So anyway, that's the long version. But you get the idea, you could get to a point where you solve these two major classes of problems in one fell swoop. And eventually, Nathan and the other folks in the kitchen started to come around a little bit. I tried to get companies like General Mills and Kraft and some of the other big food companies to look at this, and they all thought I was crazy. And so it didn't go anywhere at first, but every year it sounds a little less crazy, and now people are trying to print food.
SPENCER: I imagine there are some people that “this sounds amazing”, and others that “this sounds terrifying.”
PABLOS: Oh, yeah. Lots of people have a romantic relationship with food and think this sounds dystopian.
SPENCER: It's also interesting to think about, you could have a fixed-shaped wafer that tastes like anything. Because we usually associate the texture of food with the flavor. And the idea that you could mix and match them so you can have different textures, different flavors is an interesting concept.
PABLOS: So this is actually what people, I think, a lot of times don't understand is that flavor is solved. That comes in a bottle; you got any flavor in the world in a bottle. Aroma is solved, nutrients are solved. The only thing chefs are really doing is managing texture. I mean, they're killing trichinella, and they're killing off bacteria to make it safe by cooking food, but they're basically managing texture. And so if you think about where we're headed, (we don't have this yet) but if you think about it like on a computer, we have a computer screen that uses RGB. So we have red, green, and blue, and we can mix them to make any color we want. We don't have that for aroma. And we may never get it, but it doesn't matter. Because you can turn 56 vials of different smells, and that would probably cover the things that you usually eat. But for texture, that's something that 3D printers are really good at. We can print all different kinds of textures with a 3D printer. And it might be early 80s dot matrix texture for a while, but give us 10 or 20 years, and we'll get high-resolution printed strawberries, steaks, and French bread.
SPENCER: In terms of chefs managing texture. I imagine in real life they also manage flavor a lot with things like caramelization or braising steak (it probably changes the flavor). But I think what you're saying is that, in theory, that flavor could just be pre-prepared and added.
PABLOS: Yeah. There are artisanal ways of getting caramel flavor, but you could buy that caramel flavor in a bottle, right? (From a chemistry supplier.) That's already a solved problem. They don't do it that way, but they could. And if you look at how industrial foods are made, that's how it's all done. There's no caramelization going on in a KitKat bar. There's just caramel stuck in it.
SPENCER: So the next topic comes about the concept of “what can I make this do?”
PABLOS: That's a quote from a buddy of mine named Rick Deck, who figured out a long time ago the easy way to describe for people and explain how hackers think — I've described this a bunch of times. I basically copied his idea — which is to say, “Look, if you think about most people you know, they think in a fairly predictable pattern. If you get a new gadget and give it to your friends or your grandma, they might ask you “Well, what does this do?” And you can explain to them, “Well, this gadget is a phone or it's a camera, or it's a stereo system.” It's a very predictable question. If you give the same gadget to a hacker, the question is different. The question is, “What can I make this do?” Those are the kinds of people who are going to flip it over and take out all the screws and break it into a lot of little pieces, and then figure out what we can build from the rubble, like what is made possible by those ingredients. And I think that just that fundamental question is different. And it sort of puts a fine point on the way that different people think about the world and interact with the world. And I think it's very important to understand that because we need them. We need people who are good at violating the warranty before they've got the shrink wrap off. We need people who are good at figuring out what's possible because that is where all your new technologies come from. Nobody ever invented a new technology by reading the directions. We have to actually engage the kind of people who are not following the rules, and so I think we should cherish and appreciate those people more because they're the ones who give us our new superpowers.
SPENCER: Yeah, I'll add to what you said about this “what can I make this do” mentality. Almost all ideas are built on other ideas. So you're taking some idea and pushing it further than it's ever been pushed before to make something novel.
PABLOS: There are two major classes of ideas. The one you've described is like 99% of it or more. It's almost always cross-pollinating ideas from a couple of different areas and taking them a little further. As an inventor, even in the example I used of 3D printing food, I would just happen to be the one guy in the world who had a machine shop full of 3D printers next to a kitchen full of world-class chefs. So if you just stuck some other inventor in the same position I was in, they may have gotten to the same conclusion, which is that “Oh, we should cross these two things.” And so a lot of these things are like that. They really are. It's those ideas that come from mixing up different areas in science and technology, different scientific disciplines, different knowledge bases, and skill sets. But I also think, very occasionally, you get somebody who is truly creative, and they create something new out of whole cloth. And I think that's also very special and rare.
SPENCER: What do you think one or two good examples of that are?
PABLOS: Here's an example that I like. One of my best friends is a skateboarder named Rodney Mullen. And Rodney is like a recluse. So hardly anybody knows, but anybody who's a skateboarder knows who he is because he was kind of like the godfather of street skating. He's the one who invented everything you've ever seen a kid do on a skateboard. He did those tricks first. He was the first one to like Ollie a skateboard, which is the basis of everything. People don't realize it, but Rodney grew up in rural Florida, on a farm. He had a little patch of cement that his dad put down in the driveway to skate on. And he didn't have influences. There was nobody around him who knew more about skateboarding than me. He didn't have anybody to show him anything. He had to figure it all out from scratch, which is why he was able to figure out so many things that no one else had because all he had was his own imagination. And people don't think of those as important inventions, but I love Rodney so much and his mind as an inventor is inspiring to me because he's able to dream these things up, and figure out how to make them real. And there is no ability to connect those creations to some other influence. He came up with them on his own. And so sometimes you can see it better. Inventors are not well-documented, and in a lot of ways, but artists really are. And so sometimes you can see it more. I think people don't make a good distinction between art and craft. Most of what we see and celebrate is really craft. But occasionally you get someone creative. You don't get a lot of people who are actually creative, but when you do, it can be extraordinary. And we celebrate that sometimes in the arts. We get an occasional Jimi Hendrix. But a lot of times we don't make a distinction between how hard it is to do something the first time and by comparison, how easy it is to do it the second time, and every time after that.
SPENCER: I think about this with calculus. It feels like you could teach calculus to a high school student today, but to come up with it is just mind-bogglingly difficult.
PABLOS: Well, right. And then the history of calculus is actually a very interesting one for that very reason because it appears that Leibniz and Newton came up with it at the same time, or pretty close, even though they weren't neighbors. They lived in (what for us would be) unfathomably far away from each other and could occasionally exchange letters. So it's arguable that whatever forces are in play, or ideas in the world, or technologies that have come about kind of just converged on the point when calculus could be invented at the same time for those guys in a way that maybe it was also true like Rodney didn't invent the skateboard — people actually don't realize almost every skateboard they're riding now is one that Rodney invented — but it's a progression on a tool that existed before that. So, there are sometimes forces in play that we don't see, which maybe makes it look like things are more creative or out of the pure blue than we realize.
SPENCER: It seems like there's one type of genius that sort of constantly learns new things about different fields and tries to put them together. And there's this other type that just takes one thing and sort of disappears and does their own thing for years and years and years. And by the time they're so deep into it, they've pushed it way beyond what anyone else has ever seen. It sounds like your friend is like that.
PABLOS: Yeah. And maybe I'm like the first thing you described, so I appreciate the other one because I can't fathom it. That's probably all that's going on is just that I cheat by learning as much as I can about different things and trying to find the space in between. Whereas Rodney or whoever other types of inventors would just go deeper and deeper and deeper into their thing.
SPENCER: So on another topic, we talked about this as a podcast about ideas that matter. And you mentioned to me that one of the things in your mind is ideas that don't matter. So tell me about that.
PABLOS: Yeah. Ideas that don't matter. When I read your “ideas that matter” thing, well, I do feel a sense of frustration that when you look around in the world at what people aim their attention at, they're very fixated on a lot of things that really just don't matter that much, which is sad and disappointing because there's a lot of things that do matter that you could apply the same attention to. It's interesting because we have gotten so good at celebrating things like financial success. In the tech industry, we celebrate entrepreneurs, and we celebrate the entrepreneurs whose companies end up being worth the most. And when you look at them, a lot of them are not actually solving meaningful problems. They're just the ones that made the most money, largely by exploiting loopholes in human psychology, I suppose. [laughs] So there certainly are great and successful businesses, sometimes even good products, but no actual technology. And I think it's almost hard right now to even think about something like an iPhone. There are a lot of really hard problems that had to be solved to make that possible. There is a lot of technology in there that was long and hard to invent and develop. Think about the display on your iPhone. It is obscenely high-res. It's bright, it's beautiful, you can read it, you can see it in the sunlight. And the photographs are beautiful and amazing in a way that wasn't always true on computers. But you probably can't name a single person who worked on inventing that technology. And the same is true for the wireless radios inside. The reason that you have a phone that can talk to the internet is because there's a whole bunch of really difficult inventions that had to be made to make wireless data possible, especially fast wireless data. All of us can probably remember when our phones weren't very fast. Well, now it is. I can't name a single person who invented any of those technologies, or even name the inventions themselves because we don't celebrate them. And the same is true for the storage and the processor and all the things in your phone. You can't name a single person other than Steve Jobs who worked on a fucking iPhone, or maybe Jony Ive who picks the colors and the shape. So I find that very disappointing. But how many Kardashians can you name? You and I have no interest in Kardashians, and we could probably name at least three. So that's why I said it that way. What are the ideas that don't matter? They seem to be getting the lion's share of people's attention, and so I'm curious how we course correct on that.
SPENCER: Yeah, obviously it's a really big difficult challenge, but any additional thoughts on course correction?
PABLOS: I thought I would pick your brain about it. My initial thoughts are not highly developed. I mean, the first one is that in my lifetime, I think I gravely underestimated how important stories are to human psychology. Essentially, people seem to be powered almost exclusively by stories: stories, booze, and a little bit food. They need the stories. Those stories that are in people's minds are what drive their decision-making and their attention, their values, what they care about, and what they do and all that. And so, we got so good at telling stories. Hollywood is really good at telling stories. And it turns out, one of the things that makes a story sell better is if it's a scary story. And so, what I see happening is the world's best storytellers are all aimed at telling these scary, dystopian stories about how everything goes terribly wrong, which we must be wired to really want those stories, probably because at some point in evolution, it was important for people to know about what could go wrong, so they could avoid it. But I think we're at this stage we're so busy with consuming stories about what could go wrong, that we're really not putting a lot of effort into figuring out how to avoid it. And so, I'm interested in this because I think maybe it's one of the things that we need to start doing a more competitive job of with science and technology is how do we tell the stories about how everything could go right. Or at least how everything could go better. I've personally worked on a lot of technology projects where the goal is to make something better for humans. But the technology doesn't go anywhere because humans have chosen not to use it or not to even try it. And that seems pretty sad, where they're scared of the unintended consequences of a technology that they never even tried. So, we see that with different things that could make a difference. And so it feels disappointing because we could go invent more technology, but if it's not gonna get adopted anyway, then it doesn't really move the needle. So I don't know, what do you think? You got a podcast, which is like a storytelling machine. What's the job to do here? How do we fix humans? We're wired for stories.
SPENCER: Emerson Spartz has made the case to me — which I think is quite interesting, similar to what you were saying — that if you analyze people's time, a lot of what we do is consuming stories, even if we don't view it that way. Television is stories, books are stories, and even news articles are stories. A journalist is always saying, “Well, what's the story behind that?” There's a bunch of facts, but what's the story? Even tweets are often like little mini stories. And so I completely agree on that point that so much of what we do is consume stories. And on the evolutionary point, I think it's really interesting to consider how the story module we have is being hijacked. We think of super stimuli (like sugar is a super stimulus for food for calorie consumption) but like we have the super stimuli stories, where, presumably, the reason we're so interested in stories is they were a learning mechanism. Like, you sit around the campfire at night with your tribe 20,000 years ago, and someone tells a story about how they almost died that day because they were climbing a tree and something happened. You're listening really carefully because that's super important information. You want to learn how to not die, or someone's telling you a story about how this other person in the tribe betrayed them. And that's incredibly valuable information because you need to understand the social dynamics and who you can trust. But today, we're reading stories about the Kardashians. What the hell does that have to do with us? What do we learn? And they're super stimuli because the elements that make them interesting to the human brain are maximized because there's a structured pressure. If the Kardashian stories weren't good stories, you wouldn't have heard of the Kardashians. It's sort of like there's extreme evolutionary pressure on stories right now to make them appealing. So yeah, I think you make a lot of great points around that. I'll just add one more thing into the mix, which is, I like to think of this dichotomy of hype versus value. Where hype is all the things that make people interested in something including mechanisms of social status. Whereas value is the things that deliver what we truly care about other than social status. And so let me just give some examples to clarify. Imagine that one day, a certain style of dressing becomes really, really popular. Now, wearing that kind of clothing might deliver you social status, but other than that, there's nothing intrinsically better about that style of clothing. It's not like it protects you from the cold better or helps you achieve your goals better other than saturated goals. So I would call that pure hype, or you can have some of these cryptocurrency projects that clearly offer no value to society. But there's huge hype behind them. And people are making tons of money on them, just because they're thinking that other people are gonna buy them. And so everyone's buying and hoping that others will buy them. So that's pure hype. And then all the way on the other side, you might have an invention that literally saves people's lives. It's just like pure value. But you can't get anyone to use it because of weird bureaucratic reasons or something. So that would be pure value. And so one of the things that I've come to think about is if you can view this as like a two-axis system, these are not mutually exclusive. It's like a two-axis system, you've got hype, and you've got value. And the problem is, if you have pure value without any hype, you can't get it off the ground. It's like you were talking about, you could go invent these incredible things but nobody will use them. If you have pure hype without any value, it can do incredibly well, but what's the point? It's just a waste. It's a waste of human mindshare. And so we want things to be mostly valuable, but we need enough hype to get them over the line. So I've been thinking about that more and more.
PABLOS: Right. So I started to really appreciate that. I mean, I worked on cryptocurrency in the 90s, and we had value but there was nobody on the internet yet to use it. And it was just a bunch of nerds dorking around. And when Bitcoin started to finally catch on and get people interested in cryptocurrency, I pretty quickly lost interest because of the hype cycle, and it attracted a lot of opportunistic Ponzi schemers, and I found it distasteful. But that part, I think of it as just like a developmental stage that was also very important, because now the fact that a bunch of people did get rich off of cryptocurrency, it has attracted a whole generation of coders and users to these technologies and the crypto toolkit which could do a lot of really great things. And so now, for example, it's unfathomably easy to raise money for a project that involves a bunch of cryptography, which was never true before. And so now I have more of an appreciation for that part of the life cycle, even though I still kind of find it distasteful.
SPENCER: Yeah, I think certainly there will be some really valuable projects coming out of it, not just valuable in the money sense, but valuable in actually providing value to humankind sense. However, I think it's still mostly hype. [laughs]
PABLOS: Oh, yeah. The noise floor is extreme. I think another example of what you were describing as the hype versus value mismatch was nuclear reactors, which have, essentially, miraculous ability to provide energy, the track record of safety is bizarre — so if I remember correctly, there are currently no deaths in the US associated with nuclear power generation, which isn't true for things like even solar panels and windmills, people fall off roofs installing those things and stuff. — So there's a lot of value to be gained from nuclear reactors, which are carbon-free and scalable and provide energy when the sun ain't shining and the wind is blowing. But we regulated them into oblivion in the 80s because of those scary stories people had in their minds from the failures that Three Mile Island and Chernobyl and things. And so I just feel like it's a case where the scary stories won. These scary stories are really good scary stories. — I don't know what kind of awards they give out for TV, but the Chernobyl TV series probably got every award they could get, because it was amazing television. — But it certainly didn't do anything to help people get more excited about building nuclear reactors, which is really sad, because that's a technology that we have that exists today that can save so many lives and help with so many problems, and solve clean water, recycling, sanitation, and solve all these kinds of things that are real problems. But the stories are holding us back. So that, in my mind, is one of the big ones.
SPENCER: I've heard that the environmental movement is actually split on nuclear reactors.
PABLOS: Well, that's an improvement because we used to fight them, like in the 80s, they were dead set against it. Now, it's kind of half and half because they recognize that. I mean, at least some of them who also coincidentally have learned to do arithmetic have figured out that we don't really have any other way of getting to a carbon-free world in something measured in a few lifetimes.
SPENCER: There's this great irony insofar as this is true, that the environmental movement may have sowed the seeds of destruction by pushing so hard against nuclear early on. I think the regulation on nuclear in the US is so stringent that it's almost impossible to create a new nuclear plant. Do you know about that?
PABLOS: Yeah, I do. We invented a new type of nuclear reactor in Intellectual Ventures. It's called the TerraPower reactor, and it's a modern safe reactor that's powered by nuclear waste. We have in the US like a million metric tons of depleted uranium leftover from mostly making bombs. So our machine is a nuclear waste recycling machine, or the TerraPower reactor is, anyway, a machine that can both turn that stuff into fuel and create a lot of energy. In fact, it's probably at least an order of magnitude more efficient than today's reactors. So it makes a lot of energy. And we were unable to get the NRC which regulates nuclear in the US to let us build it for a long time. I don't have anything to do with that project anymore, to be clear, but I think that they have got some kind of provisional ability to now build a reactor in Wyoming (last I heard) so maybe things are improving.
SPENCER: It seems like now there are more and more environmentalists who are saying that nuclear is a good option. I will just add to this discussion that I think people are really afraid of some massive meltdown, so they worry. Well, okay, maybe we've had no deaths here in the US, but maybe there could just be this massive meltdown. Not only because a lot of people spread radioactive material all over the world (there's detectable trace amounts all over the world). And I think that really freaks people out. I don't know too much about the details, but my understanding is that the safety profile on these things has just improved incredibly. The meltdowns were occurring on incredibly old technology compared to what we have today.
PABLOS: Every nuclear reactor that we run in the world right now is essentially a design made with pencils and slide rulers by genius engineers before we were born. And then we've strapped on some band-aids to try and make them a little safer, most of that monitoring in our lifetime. But now we have giant supercomputers. We can model the neutron activity in the reactor core, we can understand exactly what's going to happen, and we can design modern safe reactors. So the TerraPower reactor, for example, can't melt down. It actually enriches its own fuel inside of the reactor. So if you had any kind of problem, it would just fizzle out, it wouldn't be able to have a Chernobyl-type meltdown. So all that is totally possible. Our ability to do advanced engineering has improved a lot in the last 60 years. But unfortunately, we're not working on it. The nice thing, I guess, in one sense, is that China has been very aggressively growing its nuclear reactor programs. And that's really good. Because they will probably get carbon-free long before we do, ironically, and that'll help the whole world.
SPENCER: So switching topics again, tell us about what is an innovation DMZ, by which I think you mean, innovation demilitarized zone, referring to the DMZ between North Korea and South Korea, I assume.
PABLOS: Oh, I mean, we've had different kinds of what's called a demilitarized zone, which is between countries at times. I use that as an analogy to try and describe one of the big problems I see with innovation regulatory intervention. It is the kind of thing you want to do when you understand what the risks are, what the problems are, and what you're trying to optimize. And we've gotten really, really aggressive about regulating things we don't understand yet. I think it's a huge problem for innovation. And the nuclear reactor's story actually is a good example of this, where we didn't fully understand what was possible. And we created so much regulation, and we are so much better at making laws than deleting them. And so we really pinned ourselves into a corner with that. Other examples, though, are like right now you see a lot of people up in arms about so-called artificial intelligence and how it could all go terribly wrong and how the bias that's in AI could cause exacerbation of societal problems and all these kinds of things. And so a lot of calls to regulate the use of AI and people believe that AIs are going to turn us all into their pets or robots or something. And so I think it's way too soon because this is a technology that hasn't really played out very far. We don't understand it, we don't understand the value of how it could be put to use to solve big problems, and we don't actually understand the failure modes well enough to start regulating them. That's what I believe. Another great example, though, is healthcare. In the US, we have extreme amounts of regulation around health care. It's very difficult to innovate. There is a prescribed process, but it only works for some things, and it can set back what could be very valuable innovations by decades or more. So for example, there has been an interesting progression in the last decade of legalizing marijuana, and that has been helpful for some people with different medical conditions that really weren't being served well by other drugs and things. But that also paved the way for people to be more open, apparently, to deregulating psychedelics, which as far as I know, hasn't really happened, even though it's starting to look like it could happen. But what has happened is it's allowed some people to be bold enough to take on researching what this class of drugs could do to help people. And we're seeing all kinds of extraordinary possibilities, treating things that we have no other meaningful interventions for, like treating PTSD and those kinds of things. With psychedelics (I'm not deep into that), but I think it's the kind of thing where we ought to be able to just say, “Look, if you want to experiment, then here's the place to do it.” Let people opt-in, say, if you want to experiment with psychedelics, we're not going to employ you anymore, ensure you anymore, whatever it is that the people might think are the necessary exclusions to make, but let people make their own choice. We sacrifice a lot of lives to Netflix right now. [chuckles] I mean, if you just measure human life hours or something, maybe your life expectancy was extended from age 33 to like age 73. But most of those extra years are spent watching Netflix. So why not let people choose if they want to spend some of that life, reduce their life by testing certain drugs or experimental gene therapy, or whatever they want to do? If we'd let people do it, I think we would get further faster and the net win would probably be easily justified by how many lives you save in the long run or improvements you can make to people's lives and health. So yeah, I think DMZ is kind of a way of thinking about how you could make that work. You look at what's happening now in some things, we have taboos in the US that China doesn't have. So in some sense, they're a DMZ for AI research and for genetic research, those kinds of things. So in some sense, I think that's probably a good thing.
SPENCER: So is the idea of innovation DMZ basically saying, “Okay, we're gonna have the space where people can experiment with very little regulation until we kind of figure out what this technology can do, and in what ways it's dangerous, and so on”?
PABLOS: Yeah, so for example with CRISPR. It would be illegal for me to just have you come over to my house, I sequence your DNA, and I use CRISPR to change your eye color from blue to purple or whatever because that's what we want to do. We're not allowed to do that right now. Like, I get locked up for using experimental gene therapy on my friends. [CHUCKLES] So you could say, that's not very important, changing eye color. Fine, but it's your life. It's what you want to do with your life. Maybe you decide that purple eyes are more important to you than the loss of the remaining years you might have. And you're totally willing to risk it. Maybe it's not worth it for purple eyes, but if it was for an incurable disease, why not let us try? So that's what I think is going to be important in the future is figuring out ways to keep from reducing innovation, because you see the net effect, the reason for this is that those breakthroughs will save lives. They will improve humanity going forward, and we will get them more slowly, and we will sacrifice more lives getting there by regulating against them too early. So I think DMZ is a way around that. So let a few people opt-in. We let people opt into being alcoholics and drug addicts and adrenaline junkies. I mean, I can opt into being a base jumper right now, that's not illegal, and very likely to reduce my lifespan. Why can't I opt into playing with CRISPR in my basement, or opt into playing with CRISPR in a laboratory at the research facility down the street? I should be able to do that. So that's what I think, that's probably controversial, but probably warrants discussion.
SPENCER: I largely agree that new technologies are just really difficult to regulate, so erring on the side of not regulating them too early seems good. However, I do think that there's a difference when there's a kind of Black Swan risk from technology, like for example, people experimenting with making new viruses. That seems like it just threatens the whole world, and we just have to regulate that immediately.
PABLOS: What you would do then is regulate the process of the environment or the controls around it and those kinds of things. And maybe it has to be regulated to an extreme, but that's an example of where now we understand the risk. Like if this goes terribly wrong, then it can leak out of the lab. It can spread all over the world (because of modern air travel) in like 24 hours and it will take us years to recover. Millions of lives will be lost and we know that the risk-reward ratio is high. That one, unfortunately, is one where we probably do have a regulatory environment that keeps the honest people honest. But the tools you need to do those kinds of things are — you can use a salad spinner in your kitchen and start making new viruses — so that's one that I think isn't at the stage I'm describing, which is too early to extrapolate what the real risks are, and what the benefits and downsides are. We know a lot about the risks of creating new viruses in a lab. And so some regulation is probably appropriate. I'm not arguing it's not.
SPENCER: It seems like there are certain kinds of technologies where looking at them in advance and analyzing them carefully, you can see there's going to be these kinds of tail risks, whereas others are much harder to see whether there could be any tail risk. And so I think that at least with the ones where, in advance, it's pretty obvious you're gonna have these serious risks, maybe they should be regulated a lot earlier. I would argue that viruses are one like that, where you just think about it, 20 years before we have the technology, you're like, “Oh, shit, that's gonna be bad when we have that technology to cook up new viruses easily”.
PABLOS: Right. I mean, look, that's the only thing that scares us. I'm not afraid of AI. I'm afraid of viruses for sure.
SPENCER: I'm afraid of both. [chuckles]
PABLOS: Oh, really?
SPENCER: So a sort of related topic is crypto regulation, which seems a little bit almost paradoxical, right? And this regulation seems almost like the total opposite of the crypto mentality. So what are your thoughts on that?
PABLOS: That's an interesting point because it's near and dear to my heart — not crypto as we know it now, but cryptographic protocols — what that is, is a way to create a fair interaction between people, which is the whole point of regulation. So regulation is one way of accomplishing something that is the exact same thing that crypto (cryptography) is trying to accomplish, which is how do you enforce a fair interaction between people. Cryptocurrency is just one kind of transaction or interaction between people. And it performs vastly better than what humans did before. What humans did before for centuries was double-entry bookkeeping, which is I write you a check, you give it to your bank, they increment your account, they told my bank to decrement my account, there's a whole bunch of auditors and bullshit in the middle to make sure everybody's playing fair, lots of laws, lots of regulations, lots of checks and balances. It's very expensive and adds quite a bit of drag to the system. But cryptography has done in cryptocurrency, it has eliminated the need for all of that and created an essentially instantaneous settlement of the transaction, where nobody can cheat and there are no external auditors or checks and balances. The system itself does all that. So, I think that's super valuable because a lot of what humans need to be able to do is have different kinds of interactions, and especially as we do them online and with people who we haven't got a personal relationship with, so it can't build a lot of trust. And when we might have mutually suspicious actors in a transaction, cryptography gives us a toolkit to create these kinds of relationships and transactions between ourselves. And so it's a better way of solving the same thing that you would solve with what you would call regulation. So right now, people are very concerned about this idea that maybe governments are going to regulate crypto. And what I believe is that on a long enough time horizon, there's absolutely no question, crypto will regulate governments. And the reason for that is because it's better. And humans — for all the complaints that I've made earlier in this conversation about human decision-making, being quite poor [laughs] — on a long time horizon, we usually actually do the best thing. So in the short run, we do a shitty job. But in the long run, we actually do a pretty good job. So for example, we don't often invent a technology, put the effort in to figure out how to make it work, and then choose not to do it like we have with nuclear. There is no question in my mind that in a time horizon of a century, we'll be using nuclear reactors like humans will solve this problem. Generations yet to come will have forgotten all about Chernobyl and Greenpeace. And therefore, they will move on to the point where we will build those nuclear reactors. So in a long time horizon, we win. Cryptography and using cryptographic protocols is a much better way of doing a lot of the things we are trying to do with regulatory processes now. And so on the 100-year time horizon, I expect there will be no other currencies that are not cryptocurrencies. I believe that the governance that's being used to manage cryptocurrencies now, in its infancy, will evolve and be able to provide much better possibilities for governance in all kinds of areas. And I'm excited about that because I think that's a much better future. It still relies on humans to make decisions about what we care about. We still have to define those protocols. We have to define those contracts. We have to set them up to do the things we want. But as you can see right now, because crypto is not regulated, we live in this amazing inflection point in time where we have thousands of competing projects, thousands of competing ideas, and thousands of competing tokens. And they all have different relative merits, and they're all trying to solve different problems, and some are gonna be winners, and some are gonna be losers. But there's room for all of them to coexist, and in the same way that you have thousands of companies on the stock market, and you can choose which ones of them you feel are valuable or important or likely to grow. And so I think that our understanding of currency is kind of a limited paradigm for what is possible with crypto, maybe a better way of thinking about it is contracts. We have relatively few currencies in the world, but we have a multitude of contracts, like an unlimited number of possible different contracts. And that's really where crypto goes in the future is we should think of each token as a different kind of contract. And we will have a multitude of them. We're not trying to get the whole world to stop using so many different contracts and just pick one.
PABLOS: The contract between you and your housekeeper is definitely a different thing from what Walmart needs between them and their suppliers. It doesn't make any sense to try and whittle it down to one. That is not what is actually going on with crypto. We actually need to get to a point where we support a multitude of contracts, but they're cryptographically enforced.
SPENCER: Yeah, and I largely agree that crypto can help automate a wide variety of different contracts. But I will push back a little bit in that I think there's quite a few limitations to that. A lot of contracts people want to write are not the sort of thing that, at least today, we can turn into code. And our code has to be airtight, it has to be completely unambiguous and executable. Whereas if you look at real human contracts, they very rarely are of that nature. And they very often require subjective interpretation. Maybe we'll one day live in a world where with AI, we can have contracts that have ambiguity, but AIs can still resolve them or something like this, but we're not in that world yet, at least.
PABLOS: Look, you could write a cryptographic protocol that has ambiguous variables, or variables where they come from some external system, and we use Oracles for that in current cryptosystems. And those can be as ambiguous as you want. I don't know how desirable ambiguity is in most contracts. I think that's probably a side effect of humans being unable to express or anticipate their requirements. And it isn't necessarily saying this would be the only way you do anything, You could still have ambiguous contracts adjudicated by courts and things if you want, but those are definitely more expensive. And we are having a problem right now because the courts cannot keep up with enforcing all of the contracts that humans are making. And a lot of them could be very standardized and could be implemented in a protocol. So I don't see a problem with both coexisting. This is a new tool, so you would use it where it makes sense, and use the old way where it makes sense.
SPENCER: Absolutely. I guess I see a lot of the value of contracts between people or between companies as being a declaration of what's being agreed to, rather than something that's fundamentally enforceable. The reality is our court system is so expensive and so difficult to use, that most of the time people don't go to court anyway. It's not even worth it.
PABLOS: It's failing.
SPENCER: Exactly, it's failing. BUT, the contract still serves a very important purpose, which is that it gets people to be clear on what they're agreeing to, and I think that actually has a ton of value.
PABLOS: Certainly, the most valuable thing would be for humans to get clear about what they are agreeing to or believe or want or value. I don't think we're going to come up with a computer system that can do a good job of that for us, which is too bad because we aren't very good at it. But once you get to thinking about what you just described, it's true that we write the contracts, but they're too expensive to enforce. That is a drag on an economy that is holding civilization back. You need to be able to do that in order. The more reliable my contract and my agreement with somebody can be, the more of them I can do, the more I can put on the table, the more I can risk, the more I can grow. And those things are very important to be able to take care of all these humans we made [laughs]. So that's how I think about it.
SPENCER: So let's talk about deep tech for shallow tech.
PABLOS: Shallow tech was my way of trying to describe maybe another class of ideas that don't matter. The tech industry gets a lot of antagonism these days, I guess, because people are pissed off that Facebook or Uber, or whatever is not expressing their values. And I think it's very interesting because that's who we think of as tech companies, Fang, Facebook, Apple, Netflix, and Google. What did Facebook invent? What technology have they really created?
SPENCER: The newsfeed I think? [laughs]
PABLOS: Good point, I give them the newsfeed. So it's not really a technology; it's a software. And some of it is new technology, and some of it is very useful. But I think the tech industry has really lost its way and has overwhelmingly focused on software, which you can't blame them. It's relatively easy, it's generally useful. You can apply it to anything. It's cheap and predictable to build and scale and sell and make money off of and all that. So I'm not really arguing with that. But I think we got to stop thinking of that as technology. That's the software industry. And venture capital is largely aimed at the software industry, and so it should be called Software capital. All the people making enterprise apps, we call them SAAS holes now. [laughs] That's a different thing than inventing and creating a new technology that you can bring into the world to solve a problem at a larger scale. That's what we call deep tech. These are fundamental technologies. So an example is (maybe) blockchain or the consensus mechanism that Bitcoin brought us. You could think of it as deep tech; that's a fundamental technology. Is Dogecoin? Probably not. Is the n-thousand cryptocurrency that's a minor variant of Bitcoin Ethereum technology? No, it's not. It's shallow tech. Is the next new iPhone app going to be a technology that changes the world and makes it better? Maybe they'd be a little bit, but that's software. Is the technology that's going to make 5G work shallow tech or deep tech? I'd say deep tech. If you can solve energy or clean water or eradicate malaria, maybe those are actual technologies that are very important that need to get developed and add a lot of support. And I think we don't make a very clear distinction in our minds. And so we sort of just handed the world of technology over to opportunists who are building software companies. So again, that's okay, we do need to do those things. But we shouldn't kid ourselves. If we want to make a big difference, we got to make new technologies, and the new technologies are — we call them exponential for a reason — the reason is, a new technology gives you a new exponent in your equation to solve a problem at a larger scale. And that's important because we have an ever larger scale. This is a planet that has doubled its population since I was born. And it doubled its population in the 50 years before that. We have 8 billion humans now; we've never had that many. Only 200 years ago, we had less than a billion. So imagine for people who have kids, you have a kid, you raise your kid, you saved up money to put them through college. Well, we're just kidding. You got to put eight kids through college. That's the kind of resource allocation that's necessary to take care of all these humans. Now, I don't think we're really looking at it that way. So in my mind — I don't mean to be a curmudgeon here — I just mean we need to define technologies as being very important and go after it with a vengeance. And you can't raise a generation thinking that Snapchat is technology. It's not. That's not what we're talking about.
SPENCER: Maybe we should think of it as a spectrum from shallow to deep rather than a binary. Does that seem good?
PABLOS: Yeah. I don't think people are very good at spectrums. So I think the binary is actually going to work better. But let's try. [laughs]
SPENCER: [laughs] Well, I think people need to get better.
PABLOS: Yeah. We've been saying that, but it doesn't seem to be panning out.
SPENCER: Yeah. Well, in software, you can have something that really is just a duplicate of what existed. But you could also have something that introduces a new algorithm that's never been done before and push the boundary. You could think of the consensus algorithms.
PABLOS: And that's why I describe it that way.
SPENCER: What are some of the deep tech ideas that you're most excited about right now?
PABLOS: One major class of possibilities is that humans now have sensors that can measure almost everything in the world. And we have these networks that can bring the data from those sensors back to giant supercomputers, which can analyze the data. So all that exists, and applying that to a multitude of problems is super exciting because it allows humans, for the first time really, to anticipate what the results of our actions will be and make choices that are more efficient. So we can redesign everything in the world to be more efficient. And you can kind of see this. If you look at the railroad system. At the time, we figured out that if we make enough heat, we can turn steel into whatever we wanted. And so the entire railroad system is made of steel, like all the rails are steel, all the cars are steel. And that's pretty awesome. But they're absurdly heavy. And it's just what we had to do at the time. Now, we can actually design things with efficiency in mind. You can see that in the progression of the auto industry because the car that got driven around when I was a kid by my parents was all steel, very heavy. And now you look at something like a Tesla, and it's very lightweight. It's only material where you need it. It's very efficient, and it's designed to manufacture from a material perspective. So those kinds of things are possible. And what it really means on a larger scale is now we can create computational models that allow us to really try and test all the possibilities (or a sufficient number of them) in software thousands of times before we ever do anything in the real world. And that's really unprecedented for humans. And this is one of the things that makes me very optimistic. When you go back to the things I was complaining about earlier with human decision-making, a good example of this is what we worked on in our lab by doing epidemiological models. So we were modeling, in software, the transmission of diseases in mostly developing world. This is a big problem that a lot of Americans don't pay attention to, because we don't have these problems. But you get almost a million people a year still die just from malaria. It's one disease (I don't remember what the latest COVID numbers are.) I think we're a little less than a million a year still on COVID in the US. But you look at a region like Sub-Saharan Africa, you got a million people, half of them kids under five years old, who never had a chance, die of malaria still. And that's a tough one to stomach when you start to really take a look at it. And so by making computational models, we're able to see all the factors (not all but a lot of the factors) that affect the spread of the disease. The climate, rainfall, travel of humans, the way that gets transmitted between the mosquitoes, and by making computational models that show how the disease will spread years into the future, you can then test interventions. So then we can go test. Well, what happens if we spray some DDT here or deploy some bed nets? There are some different interventions, and you can see what it's going to take to eradicate this disease. And historically, what humans would do is just as much trial and error as we could in the real world. But we might be heading on a track that just doesn't lead to any possible success. And with computational models, you can see your possible futures, and choose from among your possible futures which one you want to go for, right? And so instead of going after some fantasy future, we're going after our choice of which of the achievable futures we want. And that's the power that's possible with computational modeling. And we're extending on that every day. And it's getting better and applied to the future of companies and cities and nations and civilizations. And I think in 100 years from now, we could be in a world where humans have learned to use computers as a tool to figure out what's possible. And then really go after solving some of these big problems. And I'll give you an example from the epidemiological modeling. When the first Ebola outbreak happened, 12,000 lives were lost. And unlike COVID, that's one that managed to get under control before it spread. It wasn't as transmissive, fortunately, but it was quite deadly. In the second Ebola outbreak a couple of years later or less, only 12 lives were lost. And the reason for that, in part, is because we're able to use computational models to create optimized ring vaccination campaigns and really contain the outbreak early and optimally deploy the vaccination resources. That team has been advising 75 Different countries on how to optimize their limited vaccination resources for different diseases, and that's saving real lives. That is starting us out on a path that can help us to really eradicate some diseases. We're able to use that to eradicate polio once and for all, again, because these are things that are within our grasp now because we have these tools. But I think a lot of people don't realize that potential for them. They don't understand them and yet are comfortable with them. And so that's the kind of thing I'm excited about seeing change over the decades to come.
SPENCER: Pablos, before we wrap up, I want to ask you about patents because I know that you hold something like 70 patents. And you were at Intellectual Ventures, which is famous for having a huge number of patents. And I'm wondering, how has your view on patents changed over the years?
PABLOS: There are two issues. One is the question of whether someone should be able to own anything. A subset of that is whether they should be able to own some kind of intellectual property, which means an idea or something that could be freely copied. I own the jeans I'm wearing. If you take them for me, then I no longer have them. And so we're pretty good at understanding property rights. But when it comes to intellectual property, it's not hard to steal it. If you steal it and use it, I still have it. So what's the problem? Interestingly, we've navigated this. If you think about music, we had this problem with music where I guess it was maybe like the 40s or 50s where I could just go play the Miles Davis song on the radio for free. I'm making money off it and he isn't. And everybody believed this was bad. This is totally unfair, not cool. The songwriter should be getting paid, not the DJ. I can play at a nightclub or a restaurant. I can play anybody's music that I want and not compensate them and it was totally fine. So we fixed that with copyright law, and we fixed it with what's called ASCAP. So ASCAP is like a coalition or a labor union for musicians. What they do is they go make sure that radio stations keep track of whose songs they play, and they pay money, and that thing gets distributed to all those artists. And that's worked out pretty well. Nobody questions whether an artist should get paid for their music. Same with books and movies, and everything else. The creators should be getting paid for what they create. So a long time ago, actually, Benjamin Franklin had a big hand in creating the patent system. And patents are a way to try and solve that same problem for inventors. So if I invent a new way of solving a problem, if I invent a way of curing somebody of ALS or Alzheimer's, or if I invent a way of making a car that doesn't burn gas, or if I invent a way of getting clean water to kids in Africa, or if I invent a way of streaming music without glitches on the internet, or Netflix movies without buffering. If I invent that, the idea with a patent is that I should own that invention, and I should be able to capitalize on Nashville to make money selling that or licensing it out to people who use it. But the deal with patents, unlike copyright, is that the longest I could own it for is 20 years, usually less in practice because you have to wait years to get the patent issued by the patent office and that erodes some of the time. But the point is 10 or 15 or 20 years I could own that invention, I could make money selling it out. Now would be great for me as an inventor, because then I would be getting paid for the work I put into inventing that, and maybe I'd be able to afford to go invent something else. But the problem with this has been, I guess, partly, you probably know some musicians, you probably know some artists, maybe people who make movies and authors who write books, but you probably don't know anybody besides me whose business card says “inventor”. [laughs]
PABLOS: And so, inventors don't have ASCAP. They don't have an advocacy group. And what's happening is you have a lot of bad actors. So a lot of people have tried to abuse the patent system. And especially in America, which also became a very litigious country, you got a lot of people suing each other over a lot of bullshit. And so patents have kind of gotten a bad name, which is unfortunate. And so we tried a few different ways to do it at Intellectual Ventures. But the fundamental idea was, well, what if we could really turn an invention into a liquid capital asset that was respected that could be bought and sold and traded, and we could attract more money to inventors and get people to invest in inventors through the patent system. And so we did a couple of things. The first one, which is probably the most controversial, is we bought patents from inventors and from companies that had them, and we made a big pile of patents. And then we made money by selling them or licensing them out. So that was a way to create a market, to see the market for patents and for the inventions they represent. And that ended up being a very large scale experiment in a way that the company, I think, acquired 80,000 patents, which would be the biggest collection in the world. Half of them were junk, so we probably kept less than half of them as our portfolio. And then that ended up being controversial because people saw us as the world's largest patent troll. And the idea with patent trolling is that — which a lot of people were pissed off about because there had been companies before us that would buy a patent that a lot of people saw as superfluous. I think one of the (I don't know this case very well), but I think one of the sort of poster child cases for patent trolling was somebody had patents that they'd filed that were kind of fundamental to just doing email, and so — people don't realize like a patent examiner gets a total budget of like six hours to determine whether or not to grant a patent.
SPENCER: In theory, they're supposed to not grant patents for things that are obvious, but it's very, very hard in six hours to tell if something's obvious, especially if you're not really deep in that area, right?
PABLOS: Yeah, for sure. It's a very hard job, and unfortunately, the government turns the patent office into a profit center. So all the money that we pay prosecuting patents, which adds up to tens of thousands of dollars in fees over the life of the patent. That actually becomes profit, instead of spending it on improving the patent examiner or process and improving the quality of patents that get issued, and those kinds of things. That money gets taken out of the patent office, so that's a very frustrating problem. As a company that wants to invest in inventions, we had absolutely no interest in bad patents. We would not want to invest in or own a bad patent. But unfortunately, just the way that the patent office works, a lot of bad patents are granted. So a lot of companies are frustrated, because they're defending against lawsuits from people who have gotten hold of one of these superfluous patents, and then they're suing people for it. And so, I have some sympathy for that. But I think one of the biggest problems we had, and in the era, was that intellectual property is a kind of a game humans choose to play because it has benefits and most industries do go through a kind of a maturation process. At the beginning, when an industry is young, they often don't give a shit about inventors or where the idea came from. They just want to sell it and make money. And you can see there were bad actors, like in the auto industry, for example, this is well-documented. There was a great movie about the inventor of the automated windshield wiper, which to you and I seems obvious. We could probably invent that. We can now. At the beginning, before servos and stuff were available, it was really hard to figure out how to make that thing work and the inventor of it put a lot of nights and weekends into it to figure it out. He invented it, he patented it and then the auto industry just ripped him off. And it took something like a decade or more of lawsuits to finally get him compensated. And it's a very sad story. The point is, the auto industry learned from that and they learned and now the auto industry is really great about respecting inventors' rights. They do not infringe on patents. They buy or license the IP that they're using. And we don't really have a lot of problems within that industry. It's actually a good industry for an inventor. Same with pharma. Pharma is really good about buying and licensing IP, and patents from inventors, the toy industry is great, they learned their lesson. The toy industry never rips off an inventor, but they don't even use patents most of the time. If I invent a toy, and I go to Mattel and show it to them, they absolutely will either buy it from me or never touch it. And the reason is, they don't want the inventor of the next grade toy to hear that Mattel ripped me off. And so they're very good about that. So these are mature industries that have kind of figured out how to respect inventors. The software industry, it turns out, is the youngest and most immature of all industries. And especially over the last decade or so, we've been in a position where people in software just want to move fast and break things. And they just want to build and sell and make a lot of money and grow their stock options and go public and all that kind of shit. And they're fixated on those things. And they, along the way, have decided to throw the inventors under the bus. So one way that takes shape is they believe that software patents should not be legal. They believe that anything that you can do in software is, by definition, so easy and obvious that it shouldn't be patentable. Of course, I don't believe this is true at all, I don't agree at all. There have been definite inventions in software that are very important and valuable. And that took a lot of work for an inventor to create. And I think that should be respected. But unfortunately, we have a lot of really big publicly traded software companies that are bad actors. And so they succeeded in our lifetime. In the last decade, they succeeded at really gutting software patents in the courts. They basically invalidated software patents in the courts. And so if you are an inventor, and you invent the next thing after blockchain, or you invent the next AI algorithm that can actually drive a car, whatever you do, you're fucked because of the way that these (what we call “fangs” or whatever) the big tech companies have handled patents in the courts. And so now, there's rulings like there's one called Alice (which is worth looking at if you're interested in this stuff) that just really takes the wind out of it. And so you have almost no chance of enforcing your invention commercially with a software patent, and that's very unfortunate. It was unfortunate for us because we were heavily invested in software patents. You could think of us as a cop for inventors. And the reason is, at the time, what was happening is you had these cases, like famously I remember, Microsoft was sued by Apple for stealing their trash can icon, putting it on the desktop, which is why Windows has a recycle bin now. And what happened is, Apple sued Microsoft and said, “Oh, you're infringing on our trash can patent.” And Microsoft said, “Well, you're infringing on these 100 patents.” And then Apple said, “Oh, you're infringing on these 200 patents.” And basically what happens is at that scale, lawyers don't read patents, they just count them. And so what would happen is Apple and Microsoft ended up in a ceasefire and said, “Okay, fine, we'll stop suing you if you stop suing us.” And that's how the game is played at that level. And so what it meant is that startups were deeply affected, because now going back to the prior discussion about how courts can't keep up with enforcing contracts, well, this is a case where the courts are busy helping Microsoft and Apple do their thing. And they're really not that busy helping software companies that can be sued into oblivion by Apple or Microsoft because they don't have thousands of patents. They have three. And it's very difficult to win a judgment against a big company when you have three patents, and they have 3,000. My whole point of all this, just trying to explain to people how complex the situation is, I think there's a lot of room for improvement in patents, how they are evaluated and issued, how the patent offices manage them, how they play out in courts, how they're respected in industries, and we really need to improve that game a lot. Interestingly, the patent system was very effective and very important to the history of American industrial development. And you can see that now, it's one of the things that China has chosen to aggressively implement. We think of China as copying everything and stuff, but in the last decade, they have aggressively taken on building out a functional intellectual property system so that they can get In the same kind of benefits that we got out of having a patent system. And so I think it's a sad one where most people don't understand or respect it. You can't blame them. It's very complicated. But I can assure you, for almost anyone, it's not at all like what they imagined. So that's my short diatribe on patents. We could dig in if you want, but that's the basics.
SPENCER: I think the thing that maybe I view more negatively regarding patents than you do is that when I look at patents — I've gotten an extra kind of randomly sampled patents just out of curiosity, like, what are the typical pattern — a lot of it seems like it's not that valuable to give someone a monopoly on, like, I definitely get the justification that's someone's gonna go invent a new drug, and they're gonna have to invest hundreds and millions of dollars in developing it and testing it in order to get to the market that you better give them an incentive, otherwise they are just not going to do it. You better give them protection on that intellectual property for a long enough time that they can recoup their money and make a profit or people are just not going to do it. But when it's something where it feels like someone could invent it in a garage in a month, for me, it feels harder to justify giving them a patent on it in that if they didn't invent it, probably someone else would have. And by giving them a monopoly on it, you're actually kind of blocking people from doing things rather than encouraging them to do things. So I'm curious to hear your reaction to that.
PABLOS: First of all, I think, to invent something yourself in the garage in a month probably takes a whole lifetime to that point to prepare for.
SPENCER: Yeah, fair enough, like deep expertise and so on.
PABLOS: And so I think it is very dangerous to interpret it the way that you do and the reason is a lot of us take for granted that you go to work, you spend 40 hours sitting in your bed, staring at your excel, and you get your paycheck. But, that is not the kind of career that's going to lead to an invention. And so it's good for some things going back to what can I make this do, we need people who spend that time in the garage, and not just one month, but for years up to that point. And so I don't think it's fair to say that the work that went into this was just the work that you spent in the last month in the garage,
SPENCER: If it is something that truly like that person's unique experience allows him to create, I'm definitely more sympathetic. Whereas if it's something where like there's actually 10,000 people that could invent it...
PABLOS: Let me do it this way. Go back to the example I gave before. If Rodney Mullen makes a new skateboard trick that's never been done in all of human history, posted on YouTube in a video, two weeks later, kids in Kazakhstan are doing it better than him. Doing it the second time is really, really, really fucking easy by orders of magnitude for every invention. That's what I believe. And so for you and I, and I do this all the time, I can read a patent or see a new invention and look at that and go, “I could have done that. It's that fucking easy.” It's totally easy for us to do it after we've seen that it's possible. But creating it before you know it's possible, that is hard. And I'm not defending every patent. I feel the way you do, and I read them, like, “What the hell is this?” I invent something and it turns into 800 pages of legal crap that becomes a patent. I don't even know what I invented anymore. It's a very frustrating thing, the way it had to evolve. That said, if you read old patents, they're beautiful. Go read the old patents on toasters and read the light bulb patent, they're amazing. They're works of art. I love them. And it's unfortunate that I think they've been kind of perverted by legal necessities and a bunch of strategic machinations and stuff. But I would be much more conservative than you are about throwing patents under the bus because they seem easy.
SPENCER: Yeah, it's a fair point. It is hard, in retrospect, to tell you if something was easy to come up with. Well, one reason I'm interested in Intellectual Ventures is because it provides evidence on this sort of age-old topic, which is, what is the value of ideas. Because you hear this concept in Silicon Valley a lot that ideas aren't that valuable. That, really, it's about execution. And people go back and forth on this. And so I'm curious. Viewing Intellectual Ventures as a sort of factory for ideas but not execution, primarily, as a place where you coalesce ideas, and then you can sell them and so on. What do you think this teaches us about the value of ideas?
PABLOS: What I think is that ideas have to go through a lifecycle. And the lifecycle is just like, look, if you're gonna make a human, you have to meet somebody, has to breastfeed it and put it through school and get it into college, and then get its first job and get it out of the house and all that stuff. That's how you do it. But somebody has to do conception at the beginning, which I think is the fun part. But other people like the other parts, and I think we're overly fixated on the execution part, which is maybe like, get it into college part or get it its first job part or get its stock options, and all that kind of stuff. We're very fixated on those things in the commercial success and stuff. But I don't think we're very good at tying all those connections back to the invention part at the beginning. And so I feel motivated to kind of advocate for that and try and improve it. And I do think if you just take a step back and look at the technology industry, and ask yourself, “How many good ideas is this industry delivering?” You're so fucking focused on execution and finding the next great entrepreneur, well, how about a stretch goal of having them deliver a good idea to the market while they're at it? In my mind, I don't think the hit rate is that great. Despite all the economic activity, what I would like to see is more actual technology developed. And that's what I work on and what I try to support.
SPENCER: I think it's a really good point. Because, in some sense, ideas are cheap because there's so many derivative ideas out there. But truly novel ideas actually are maybe way undervalued. I think a lot of people don't realize how rare they are. How many ideas that we see are just slight variations on ideas that have been around for 30 years and have been tried so many times in so many forums, but there's like truly new novel ideas. Actually, we don't have any of them. We need to incentivize them.
PABLOS: I don't value the ideas just because they're novel. I value them for their ability to improve things and solve problems and make a difference. And so if you just start from that point, I mean, I love creating technology, and I think it's very important and I'm not trying to throw the whole tech industry under the bus. What I think is we celebrated entrepreneurs. We support them, we fund them, we back them. And that's great. But along the way, we threw the inventors under the bus and we didn't equally celebrate them and appreciate them. Because their contribution might have only been that flash of genius or that one idea or invention that took them a lifetime to come up with. But they didn't manage to turn into a functional middle manager and stick around till the IPO. So they didn't get rich, they didn't become the poster child on the front of Forbes. So I'm being a little cynical about all this. But I'm trying to make a point, which is that the world is really missing out by not creating more support for these folks. Intellectual Ventures was one way of trying to go after that. It had some success, but we certainly didn't get as far as I would like to have in backing in ventures.
SPENCER: I have to say, I'm much more excited about what you're working on now, which is trying to get more deep, deep tech built. I think that's incredibly important.
PABLOS: Now, I get a chance to not only do it with Intellectual Ventures. It should be clear, I was on the in-house invention team, and I was lucky to be able to do that. And we became prolific inventors on our own inventions but pre of the commercial responsibility that most people have. Now I get to support outside inventors, I fund mad scientists. So whoever's got like crazy hair and a DeLorean send them my way.
SPENCER: That's perfect. So do I talk about your podcast or what you're doing now?
PABLOS: Let's do it. What I'm doing now is influenced heavily by the work of Intellectual Ventures. I'm trying to, in some sense, co-opt the machinery of venture capital and aim it back at developing new technologies and new inventions and things. And that, as you can tell from listening to me, I think is too rare but solvable. And so I have a venture fund, and I back those kinds of founders. And then, the podcast is really just a way for me to share these conversations that I have. I'm trying to always learn about the biggest problems in the world, and trying to learn about the technologies that could help us solve some of them. And so I basically spend my life having conversations with smart and interesting people that I could find like you're doing, and I started to share them on the podcast sometimes because I think a lot of people aren't gonna get access to all the people that I do. And they don't get to have those conversations necessarily. So I'm hoping that that'll just help the people that are out there that want to find a way to learn more and do the same kinds of projects and advanced technology.
SPENCER: What's the best way to find the podcast?
PABLOS: The podcast is called Deep Future, and it's everywhere. That is also the name of my venture firm. And the website is called deepfuture.tech. So the podcast is there and all the other stuff.
SPENCER: Fantastic. Pablos, this was super fun. Thanks so much for coming on.
PABLOS: Oh, man. Thanks, Spencer, it's been awesome.
JOSH: A listener asks, what tips do you have for someone studying math?
SPENCER: I think it depends a lot on your goals. If your goals are merely to get a degree, that's one thing. It's just like studying for the test. But if you actually have a goal of deeply understanding math, I think one of the best things you can do is to try to figure out how to use math in the problem that you actually care about solving. So you kind of learn math while you're trying to solve some problem in the world. And so you're kind of seeing the connection between those two things. Another thing that I think is just incredibly valuable is if you know someone who knows math, well just have conversations with them about the topic. And if they enjoy talking about it, they enjoy teaching things. I think that's an incredible way to learn, to talk to someone who really knows a subject deeply and just ask them a lot of questions and go back and forth that way. And I feel like for topics that you have been really confused about, that's often been the best way to resolve them.
Click here to return to the list of all episodes.
Sign up to receive one helpful idea and one brand-new podcast episode each week!
Subscribe via RSS or through one of these platforms:
Host / Director