Top Menu
Menu
Podcast

Artificial Intelligence and the Future with Kyle Russell

By June 27, 2018 No Comments

 “That’s the scary output of AI in the near term. It’s not necessarily a killer robot. It’s humanity inflicting the worst parts of ourselves back at ourselves.” — Kyle Russell

Dr. Parsley’s sleep remedy was designed to help Navy SEALs to overcome some of the sleep challenges that they have as hard-charging individuals. Doc Parsley believes that proper sleep and recovery is absolutely essential to maintain our ability to perform at a high level. His sleep “cocktail” includes a number of supplements to provide our bodies with chemicals naturally produced by the brain to encourage sleep. Commander Divine is a huge fan and encourages members his tribe to try it out for themselves. Enter “unbeatablemind” at the checkout on www.docparsley.com  to get 10% off.

Kyle Russell (@kylebrusell) is a prodigy, having started work with the Silicon Valley venture capital firm of Andreessen Horowitz at the age of 22. He talks with Commander Divine today about the future of AI and the importance of understanding how to correctly use technology as a tool.

Hear about:

  • How artificial intelligence is already involved in our use of online platforms like Facebook and Google
  • The possibility that because our intelligence is part of our body, artificial intelligence may also need to be “embodied”
  • The unique way that the founder of Bitcoin is anonymous and has stepped away from the project to let others continue with the development

Listen to this episode to hear a very unique take on the future of the economy and artificial intelligence

PowerDot is an electrical stimulation device that allows you to increase performance, speed up recovery and overall achieve a deeper mind/body connection. Many stim devices can be clumsy and hard to use. PowerDot achieves simplicity and is well-designed. They put professional level physical therapy in your hands easily and inexpensively.

PowerDot loves the SEALFIT and Unbeatable Mission, and have generously offered the tribe 25% off of their device. You can check it out at powerdot.com–use the code “UnbeatableMind” at checkout and receive 25% off one of my favorite tools for achieving increased muscle performance and recovery.

Love the Unbeatable Mind Podcast? Click here to subscribe on iTunes.

We’d love your feedback, please leave a rating and review.

Transcript

Welcome back folks. This is Mark Divine with the Unbeatable Mind podcast. Thanks so much for your time today. Today I am at Dodger Stadium, and we have kind of a really interesting confluence of events. I came up to interview Spartan founder Joe Desena. And then he interviewed me. And then we interviewed someone else. And now I’m interviewing someone else. And I may interview someone else after that.

Anyways, before I introduce my next guest, Kyle Russell, let me remind you that my book “The Way of the SEAL,” the 5th anniversary edition is due out on Memorial Day. You can pre-order it or take a look at it at wayoftheseal.com. And if you do pre-order it at that URL we’re offering you a free workbook PDF, which has all the exercises and cool things in it. You can download from that. So wayoftheseal.com.

Also a reminder Burpees for Vets challenge. I really need help. We’re doing 22 million burpees this year. Raising a minimum $250,000 for veterans who are suffering from Post-Traumatic Stress. 22 vets a day are committing suicide. It’s unsat. We gotta do something for them.

So we’re going to do our part but I need help. So go to burpeesforvets.com… that’s hard to say all one word, fast… and you can learn how to support me or set up a team to come help.

Introduction

01:44

So Kyle Russell. Thanks for doing this Kyle.

Kyle Russell: It’s my pleasure.

Mark: I dragged you in here. I dragged Kyle in because some of you that have been listening to me a long time, know that I’m just really interested in technology and future confluence. Singularity. And also what it means to be human in this accelerating world. In fact, the first chapter of the book that I’ve just re-edited or I’m reintroducing–it’s called “Leading in Accelerating Times.” This whole perception that technology is making things go faster. Even though that’s just a perception, it’s not reality, right?

Kyle: Yeah, but our entire experience of reality is via perception, so…

Mark: Via perception. Right.

So Kyle before we get into some interesting things… like Kyle talks circles around me. Kyle works with a company called Skydio and they make autonomous robots. And he was showing me this super-light, fully autonomous drone that weighed 5 ounces.

Kyle: 2 pounds.

Mark: (laughing) To me it felt like 5 ounces.

2 pounds? Really?

Kyle: Yeah, so without the battery it’s a little bit lighter, but even with… We’re using really advanced aerospace grade materials. Anodized aluminum. Carbon Fiber. Still comes in at 2 pounds.

Mark: So this device… so you program it. So you program to do a specific thing, and then it goes and does it. That’s what you mean by fully autonomous? It’s not like, thinking, “Hey, now I need to go over here,” on its own. It’s not AI.

Kyle: It’s actually closer to the AI side than the kind of naive just go over there. So the way most drone work when you have them do an autonomous operation, you essentially… Let’s say you were trying to do a capture of a building. You want to generate a 3D model so you can inspect it for damage after a hailstorm.

Mark: Captured by video. Not like a Navy SEAL team capturing…

Kyle: Exactly. You’re getting imagery of something. And so let’s say you’re looking at a rooftop and you want to look for hail damage, because you’re doing a job for an insurance provider. Basically maps that pilots use, you kind of draw a box around the building, it creates this kind of naive waypoint path of altitude and GPS locations. And then it just kind of follows this.

But it’s not really aware of the environment around it. And so what happens–a lot of people who use drones regularly run into this–they crash. Either from user error, or just the conditions changed and the drone wasn’t able to handle it.

And so Skydio R1 built from the ground up to see and understand the world around it. And actually be able to act on that understanding. So it’s got 13 cameras…

Mark: So it’ll fly around the antenna that nobody knows is there.

Kyle: Exactly right. Even with… so let’s say you’re trying to do a cell tower… Other drones they’re kind of screwed up because of electro-magnetic interference coming from these towers.

Ours doesn’t need to communicate with the outside world to do its job. It can see the world. It’s using Deep Learning. I won’t use too much jargon, but basically it recognizes what things are and based on what it sees, that changes its behavior. So it knows how to act appropriately given the context.

Mark: So here’s the obvious question somebody’s going to ask. What if some bad guy hacks into the controlling mechanism and just changes the mind to recognize Mark Divine as a bad actor and deliver some sort of payload.

Kyle: So there’s a couple of things that we’re kind of actively doing to prevent bad actors. One is we don’t offer that level of access to kind of define at a lower level. Like, this person is bad. Or kind of change it’s understanding of the world.

The way that… the kind of access we give to folks that let’s say want to figure out how to use it in inspecting a construction site to see how well it’s going. Are they ahead of schedule or behind? Or what’s going on?

You basically give it missions, where you’re saying here’s roughly what we need to get imagery of. And in real time it’s going, “Okay, here’s all the things I need to avoid. Here’s the path I need to follow in order to get the shots I need. And dynamically, if a person comes into my path, stop and let them pass. Or go around them.

Mark: But surely the military has been knocking and or you have competitors that are doing military versions. And our Chinese friends, and our friends in other countries are working on the same technology. It’s pretty terrifying when you think of autonomous drone swarms…

Kyle: So we’re pretty far ahead of our peers. We’ve got a lot of inbound interest from industry, government, Can’t speak to how far along any of these conversations are, but there’s plenty of interest to basically. One thing that people want to use drones for is to take humans out of harm’s way. And so you can imagine going into locations where you’re not sure, is there risk or not?

Mark: Like a FEMA type environment?

Kyle: FEMA type environment. Military operations where you have human lives at stake and you don’t want to risk sending them somewhere where they’re going to put at risk.

But going back to the bad actor problem. So one, there’s a certain level of control we still have, where we can define missions, but we’re still defining the underlying objectives of like, “You’re not hitting things and you’re just capturing visual imagery.”

And with regards to, “Oh, what if the drone carried a dangerous payload.” We actually detect weight. The pressure applied to our drone. And if someone has attached anything that isn’t supposed to be there, we actually won’t let the drone fly.

And so we’re kind of taking active steps in a bunch of little ways to undermine the tricks someone might try to pull to use our vehicle in a way that we didn’t intend.

Mark: You know, I probably should have said, when I was introducing you about Kyle. So Kyle’s like a… I heard this word earlier… but like a child prodigy in technology. He actually was hired by Andreessen?

Kyle: Yes.

Mark: You were 19?

Kyle: I was going on 22 when I was hired by Andreessen.

Mark: Andreessen’s one of the top tech VCs.

Kyle: Yeah. So Mark Andreessen. One of the co-founders of Andreessen-Horowitz. He invented the web-browser back in 1993. Actually a month before I was born, they released the 1.0 of Mosaic. Mosaic which became Netscape, that’s right.

Mark: So you were brought in to help kind of advise and think about technology, emerging technology.

Kyle: So not just advise. I was actually a member of the investment team. So the kind of areas that I was looking at, we consider them emerging computing platforms. And so these were things like augmenting virtual reality. Drones and Robotics. AI. Quantum computing. Cryptocurrencies and block-chain. So basically the things that have a credible path in front of them of becoming the next big thing. Whether it’s AR and VR maybe being the step after mobile–what are the computers everyone has on them at all times? And how we engage with the world. To Quantum computing, like what’s the future of the cloud and the big infrastructure that we run really big, advanced applications on. The power and infrastructure that big companies use.

So looking at it from different perspectives, and trying to find what companies even have a credible path to that.

Big Ideas

08:37

Mark: Right, right. That’s pretty fascinating. So let’s talk about some of the big ideas in some of those tech fields, right?

First, let’s just start with AI. Cause everyone… the movie industry has been pumping out all these different views of the future with AI.

Kyle: That have a little bit of a dystopian tinge…

Mark: Yeah. There’s this dystopian tinge. To be frank, it’s pretty brutal. And so the question is in your mind can AI, like in a Kurzweilian future world, gain sentience or have an awareness that… obviously a different type of intelligence than a human. But basically be an autonomous, sentient thing. Like we see in the movies.

Or is it more of a practical tool.

Kyle: So I think the fact that we don’t really understand how consciousness works for humans, makes it really hard for us to evaluate the actual timeframe to developing a comparable system artificially.

I think that what we’re getting at is the ability to act in very human ways. The problem is that people are really good at anthropomorphizing things. We look at our dog, and we go, “Look at all these things they’re doing that are clearly have intention behind it. It’s acting like a person. It’s thinking about its actions.”

And so we’re going to see things driven by deep learning or reinforcement learning that the behavior, the actions that robots or software agents are able to take looks increasingly like human equivalent in terms of like, “Wow, that output was impressive.”

Mark: And far superior to the human capacity…

Kyle: A lot of the same advantages as just like computers generally. They can do things a billion times in the timeframe it would take us to do it once. One iteration.

Or do one task many, many times. And really quickly. And so I think that there is the side of things where are there certain classes of jobs that are going to be done by robots where they were previously done by people and maybe those jobs were fairly low-skilled labor. But I’m not so much worried about that transitioning at some point. Like in an unknown timeframe, flipping to and now we have conscious robots. And

Mark: (laughing) And humans are pets, right?

Kyle: Yeah. I think that that’s something where we’re talking more decades than you can count on one hand. And maybe even a century out. Although one of the things that’s kind of…

Mark: But it sounds like you think that’s a possibility.

Kyle: Yeah. And one of the things that also makes it… I don’t want to put the date too far out is that advancements we make on several fronts. Whether it’s the computing hardware itself. The underlying techniques that come out of academia about how these systems work can accelerate progress much faster than we expect. And this is where something like deep learning in the ’80s, people were conceiving of these neural networks. In the 2000s they had the right idea…

Mark: This concept known as the “breakout.” So the breakout is from my understanding is basically when whatever the AI project is–cause there’s a lot of different types of AI–and you know this better than I do obviously. One project or one strand would be to try to replicate the human brain through imaging.

Another strand would be through quantum computing and trying to basically create a super-intelligent being like that.

But one of these might work, and the breakout moment is when the self-learning happens at an accelerating rate to where it’s just suddenly you have an interesting thing, and all of a sudden you have the computing power of the entire human race. And it all happens, like, in a day.

Kyle: So the path that feels most likely to have a potential “breakout” like that would be… there’s a kind of subset within machine learning academia where they’re essentially making neural nets that can come up with new neural net architectures. That you’re constantly optimizing to a better architecture that accomplished the task that you want in a more efficient or faster way.

And that’s something where you could imagine… it’s very easy to go into a sci-fi… what’s the crazy version of this? What’s the crazy outcome?

Where it accidentally generated consciousness for itself, because it created a sufficiently advanced architecture. And we didn’t know what that architecture would look like, but by this thing essentially running billions of variations and different approaches on its own. Just kind of stumbles upon it.

Mark: It becomes self-aware…

Kyle: Monkeys use typewriters, but instead of typing text, it’s coding.

Mark: Right. Interesting.

Yeah. That’s a fascinating subject. Again, this is… that makes sense if you believe that consciousness is complexity. Which is Ray Kurzweil’s whole argument. It’s complexity of the human mind and the interactions that lead us to be self-aware or conscious.

I have a very spiritual training through the martial arts, and through yoga and meditation. And I don’t really agree with that approach, cause my experience is that consciousness extends beyond the brain. That there’s a field of awareness that includes our interaction with the world and our entire nervous system. Our heart and our belly. And research on the heart and the biome are now saying that…

Kyle: Thinking along the lines of the embodied consciousness and the fact that it’s not just this brain in a vat.

Mark: It’s not just the brain in a vat. The organ of the brain.

Not to say, though, that the human body/being couldn’t also be replicated as a brain or could be…

Kyle: Right. That’s where it’s like, “Oh, will you give it touch sensors? And the ability to detect heat? And the ability to actually act on the things…?”

Mark: So there’s some emotional AI, and there’s some research into that.

But then there’s the spiritual aspect. It’s like, “What is the energy that’s infusing it? Is it just electricity or is there intelligence in the life-force, the Prana. The eastern concept of Chi life-force or Prana.

And how does that affect consciousness. Those things haven’t been looked at because those are taboo subjects for the scientific community. And so they just like, “Well, that’s just voodoo. That’s metaphysics. That’s beyond physics.

Kyle: But they’re areas where when you don’t know what you’d be measuring or testing, it’s hard to have a grounded theory. And that’s where, again, we might just find like new perception capabilities or again, maybe it’s like throwing AI at the problem actually lets us find, like, “Oh, these are the observable things connected to this.”

Mark: The AI finds the solution to its own consciousness.

Kyle: AI discovers consciousness. Solves it for us. Then we can make one for them.

Blockchain and Cryptocurrency

17:49

Mark: Do you see AI as kind of a partnership… like, a handshake between this human created intelligence and humanity? Like is your vision of the future that there’s this kind of like copacetic partnership? Or more the dystopian view that it could turn into an antagonistic or both. There’s good and bad in both.

Kyle: So good and bad. But the way I think about is for the foreseeable future, machine learning in its various forms… based on different techniques… they’re all going to be built into tools. And just as any other tool technology can be used for good and bad ends, it will be misused. And so when people…

It’s fascinating to see things like Mark Zuckerberg being dragged in front of Congress to testify. And then people going, “What do you think?” Asking him about, Elon Musk style…

Mark: He’s like, “I just build a tool to connect.”

Kyle: Well, no. But thinking about “What do you think about the possible ramifications of AI?”

Well, we know the possible ramifications of AI. It’s actually here today in the form of a news feed that’s trying to optimize for sending you content that you find appealing. And kind of activates your confirmation bias parts of your brain.

It turns out it generates hyper-filter bubbles where we all live in different information realities.

That’s the scary output of AI. In the near term. It’s not necessarily a killer robot, its humanity inflicting the worst parts of ourselves back at ourselves.

Mark: I love that. That’s true. Interesting.

Okay, so we covered AI. Obviously a lot more to cover, cause AI is embedded in all these other technologies.

What about the blockchain? I laugh cause a few years ago I was all into Bitcoin. I bought a bunch of it. And after Mt Gox got hacked I sold all but 5. (laughing) And I’m like… my wife and I were laughing cause I’m like, “Sorry, Honey.”

Kyle: Sorry for the 2nd and 3rd houses we no longer have.

Mark: Exactly. “Sorry about that. But I still have 5.”

Anyway, so I’ve been really starting… I said, “Okay. That was a mistake. I really got to investigate it. Cause this really is transformative tech that’s going to affect all industry probably at some level or most.

And so I’ve been looking into some of the projects, some of the… like Neo and EOS and Bitshares and some of the really cool blockchain projects. What’s your familiarity with blockchain and what’s your view on how this is going to… distribute a ledger technology in blockchain is going to change the way… decentralize things and change the way things are done.

Cause it’s not just the currency… there is the currency thing. But this is way beyond.

Kyle: So I think it’s also… I don’t want to dismiss currencies because it’s actually going to play an important part. So a little bit of context–Andreessen-Horowitz they were very early… like, Silicon Valley venture capital firm with an actually reputable name going pretty heavily into the space. So…

Mark: 3 years ago, right?

Kyle: More than that. So back before it was hot. Chris Dickson, the general partner that I worked with… he’s actually a board member over at Coinbase. And made that investment kind of right around…

Mark: Amazing how fast Coinbase has grown.

Kyle: Oh they went from, like, being considered kind of a joke in just overall Silicon Valley. Like, “Oh, the whole Bitcoin thing’s going away. Who needs a Bitcoin bank if no one’s going to be using it? And that’s all going away?”

And now they’re bringing in, ungodly sums of revenue. And very quickly.

This is not insider information, but if you just look at the hiring they’re doing and how much revenue they’re clearly bringing in, they are on the path to be a giant public company at some point.

So blockchains, cryptocurrencies, crypto-computing… however you want to think about that. Couple different angles to it. A lot of areas I talk about, I say I’m of two minds. Cryptos an area where I say I’m like of 3 or 4 minds.

So I’ll start with the positive. So on the positives, the kind of eco-system that’s been developing around these technologies… starting with Bitcoin then kind of ramping up. Especially with the launch of Ethereum. And other platforms that have tried to be not just this ledger of transactions that have happened. That you can’t double spend and you can kind of have this entire monetary system that’s trustless and decentralized.

All right, now let’s also run computation on top of this. Or store things on one of these decentralized networks. Or do large-scale computation.

Mark: Run apps on top of it and have a whole ecosystem.

Kyle: So what’s fascinating is–and this is like a very… we talked about this at Andreessen-Horowitz… a Dicksonian theory of the world–after Chris Dickson. The idea that this is kind of the first potential force that could enable kind of a swing back away from centralization in big technology.

Right now a ton of power has accumulated in the hands of the FAANGs–Facebook, Apple, Amazon, Netflix, Google. If you look for instance the ad industry…

Mark: The power or the control of information?

Kyle: That’s one way. There’s multiple ends. You can look at essentially monopolistic behavior. But there’s also monopolistic behavior in terms of just like creating reality bubbles. Thing I mentioned earlier. Who says what truth is?

Where basically since the mid-2000s concentration in this handful of players that matter to the detriment of frankly the rest of the Silicon Valley ecosystem even. There’s some upside in that… Okay, if you have any modicum of success you can assume that like, Facebook scooping you up is like maybe an option.

It’s like, “Oh, they’ll just buy you up for your talent. And take you out. Make you not a competitor anymore.

And that’s like kind of a good thing if you’re someone who needs to actually get money back to your venture capitalist investors. Or be able to afford a home in Silicon Valley.

But at the same time, having only a handful of companies have that much power has a ton of downsides.

And so having the potential to make all these applications that we have that make the modern world as impressive as they are. The fact that we have these phones in our pockets that give us access to all of the world’s information and the ability to communicate with our friends, family and strangers who have similar interests very seamlessly. That’s amazing.

But imagine if that communication all happened in a way that didn’t lead to concentrated outcomes.

Mark: Information was decentralized. You owned it. You could…

Kyle: Or, yeah you own your information or you at least have greater control over what other folks can do with it. Making a more intentional trade-off of “You can have my information, but you have to provide me some value.”

Mark: Most people don’t understand how that works. So let’s compare Steemit to Facebook. Help the listener understand how is Steemit social platform different than Facebook. So what’s the…?

Kyle: Steemit basically combined underlying social platform that looks… you want to compare it to Facebook. I’ll get to that. But it’s heavily inspired by Reddit. Where people submit content and then using a function of either being able to up vote it. Say I like this piece of content. Or down vote. Kind of filter out as a community…

Mark: So fake news gets down voted theoretically…

Kyle: Presumably. We find that people like sharing things that confirm their biases whether it’s true or not. And so these systems will have the same problems that non-decentralized systems have in that human nature is pretty awful at some of these things. We’re just not good at acting rationally.

Mark: But there’s no central actor that owns the data. But there is a company called Steemit, right? Somebody runs the organization.

Kyle: Right. And so this is an interesting thing where for a lot of these projects the idea is they kind of start off centralized, they raise funds from the community by selling some kind of token associated with the network that they’re going to operate.

Mark: Initial coin offering that the SEC is cracking down on.

Kyle: Right. And people are going to try and come up with new names for this concept to try and get around that. Or there’ll be multi-stage processes where they only work with private investors and then open up to the broader investing public.

The idea is, you end up funding the development of this platform in a decentralized way. In that anyone that’s interested in the idea and believes it could be valuable can contribute fund to it. And have it be developed.

Then you have this concentrated group of developers who have bought into the vision, are working for the company that will get it off the ground. Make the core parts that will actually be a functional application.

But then, because you have these tokens out there associated with some kind of value being created on the network, you can actually decentralize the entire operation behind it.

Mark: So the company can step away or dissolve…

Kyle: Right. And this is where I’m actually going to bring it back to Bitcoin. Probably the most interesting thing that happened around it is that Satoshi, the anonymous creator, stepped away from the project. He’s not like, a billionaire who goes to a bunch of conferences and says, “I created Bitcoin.”

Mark: Do you think Craig Wright is Satoshi?

Kyle: Not really. I don’t have a particularly informed opinion. Like, I don’t necessarily have insight into okay, but who is?

I would maybe believe even also arguments that it’s actually a group of people. Who knows? But the fact that this founder is so counter-Silicon Valley. The founder who is able to step away and this group of people who had incentive because they owned these tokens–Bitcoin. That powered the underlying Bitcoin network. More incentive too… these things will become more valuable if we make the underlying infrastructure more scalable, faster, more secure.

And so we are going to develop it. And it’s the same dynamic that comes with working in a start-up and having equity. And wanting to make that company as good as possible, so that the equity is worthwhile.

But it ends up not actually happening. This kind of top-down company led by a CEO defining all of those…

Mark: And that’s the big argument against Ripple. Or XRP. Is that it is a company and they intend to be centralized and run as a company. And yet they’re also pretending to be a digital currency. So it’s interesting to see.

Kyle: And there’s also pseudo-transparency around whether XRP–the token associated with Ripple–is actually… when Ripple the company has clients, are those banking clients actually doing anything on top of the crypto-currency? Should its value be associated with their progress going to market?

Looks like no, from the actual information we have available.

Mark: You know, this might be just way too in the weeds for the average listener, but what happens if the SEC says that Ethereum and Ripple are securities?

Kyle: So the tricky thing is the cat is already out of the bag with a lot of these things. And what I mean by that is once you own Ether–the token associated with the Ethereum network–yes, many people who are–especially the people who are outright speculators. They have them in an account like Coinbase where the SEC can tell Coinbase, “Actually, we’re shutting this all down. And these people don’t own this.” And crazy things could happen around that.

But for the hardcore folks. The folks who again are not just speculators, but actively trying…

Mark: Trying to build apps on top…

Kyle: They’ll probably move their Ether to a cryptographically secure wallet that’s not accessible from any service and therefore if the SEC says they can’t have it, there’s not actually the SEC can do to act on it.

Mark: Right. Cause they can… it’s traded worldwide. That’s the thing a lot of people don’t realize. Even the SEC maybe. It’s like, you can say something, but I can still go trade on Binance or some other… you know?

Kyle: And that’s where a lot of the outcomes with this whole ecosystem are going to be heavily influenced by regulatory arbitrage. Where can I make the most progress on this idea I have without the government stepping on me?

And so…

Mark: Yeah. Binance was going to move to Hong Kong. Wasn’t it Hong Kong or somewhere?

Kyle: Something along those lines.

Mark: And then all of a sudden the regulators were “blah, blah, blah. We need this and that and the other thing.”

And they’re like, “Screw it.” So we’re going to go to like Macao or some crazy little country who said, “Please, bring us your billion dollar organization.”

Kyle: And that’s where you see countries like Estonia who are pushing forward on a bunch of crazy projects involving having a digital ledger for their identity and health information and all those kinds of things.

There’s some places that are going to embrace it. And that’s where people shouldn’t expect the hubs of this industry to be the same hubs as either finance or technology today.

Mark: One of the projects that I invested in is called “Bitnation.”And they’re a basically a decentralized government platform. And so there are projects… this is the good use scenario. If you’re a Syrian refugee, Syria’s not helping you out. Nobody want you. There’s is an organization where you can become a citizen of a digital country that’ll provide you legal services, provide you insurance. Provide you notary services. Now it’s virtual stuff. It’s not going to provide you a home, because there’s no geography. Although there are some countries like Estonia that will also have a digital, blockchain based currency in partnership with someone like Bitnation.

Do you see like a global citizenry evolving that transcends the geopolitical borders of country-states, nation/states?

Kyle: So this is… there’s a subset of the community involved in crypto… I’m sorry if any of the listeners here have enough of an opinion on the use of “Crypto” to be made that I’m using it to talk about cryptocurrencies and not cryptography. But that’s something that people get very mad about.

But there’s a certain subset in this community that believes in the sovereign individual movement. The idea that with the rise of the Internet but especially blockchains, distributed ledgers that let you have money that isn’t under government control. Have your information not be owned by companies or governments. And freely share information without censorship/

That the idea of having all of your rights and identity and all that kind of owned by a government based on where you were born? That becomes absurd.

An idea that you should be able to freely move between places based on where labor conditions are best for what you want to do. You shouldn’t necessarily be forced to pay into one social security system… again, based on where you’re born, but maybe based on which system do you want to opt into that you think actually accomplishes the ends that you want. That works in the way that you want.

The interesting thing there is Nation-states aren’t going to go down without a fight.

Mark: True that, right. True that.

Kyle: And so that’s something where again, I don’t feel particularly confident going what will happen. I think that there’s going to be tension between the fact that technology now allows this discussion to even be a possibility. And Nation-states wanting to still have reliable tax revenues and to offer services to their communities. And not have this weird, multi-tiered system where some people are able to opt out because their just like smarter about these technologies…

Mark: Or mad at the latest president…

Kyle: Right. But it does offer really interesting opportunities. You could imagine entirely different systems where legislation and social safety nets are essentially opt-in. And you basically can… and this is where again, you can imagine two sides of the coin. You can choose which nationality is your tribe. And again, opt-into it.

At the same time, that means you can choose your tribe and opt into it at the exclusion of others. Which is, again, very similar to the problems that we’re having…

Mark: Back to the self-reinforcing bubble.

Kyle: Right. Exactly.

Mark: I see… my more utopian view is there’s anvil. So you have your geographic citizenship, but you might have several citizenships. Which is not dissimilar to an interest group on Facebook or something, but there’s real rights and legal obligations that are recognized at an international level that go with those.

As opposed to just a Twitter feed or a Facebook feed.

Kyle: And an element there is opt-in will come faster than opt-out. We are going to be able to choose this distributed network acts as a universal basic income for people who opt-in. I can choose to contribute so that other people who want to be able to take time off of a traditional job can training and learn and go into some other field.

You’re going to be able to opt-into those kinds of systems really easily, but you’re not going to any time soon be able to say, “And I’m not going to pay my taxes to Uncle Sam to be able to fund the UBI.”

Mark: Cause they’ll send someone knocking on your door. Really quickly for that. Okay, I love that.

Use the Tool, Don’t be the Tool

36:25

Mark: All right. So enough geeking out on this. Let’s go back to the basic premise of Unbeatable Mind is how do we live a really good life and navigate our way through all these changes. How do we optimize ourselves physically, mentally and emotionally? And even spiritually? Whether working within like you are, or in spite of the technology?

What’s your view on how an average citizen is best suited or can best leverage advanced technology without losing their sense of humanity or their sense of self.

Some people go all-in. Especially the younger generation and it just seems like they get really unhealthy and really distracted…

Kyle: So I think the key is to use it as a tool, not become the tool. And this… when I say become a tool, I mean, you’ll get folks… let’s say social media addiction. It is really easy to go after the social media companies isn’t it? But again, this idea that we’ve provided these frameworks around ourselves that incentivize putting out views of the world or information about yourself that puts yourself in a best light within a certain audience. And it leads to teenagers thinking that they are missing out on all of the social fun and need to put on the presentation of being the coolest person in their community in a way that like, social pressure was constrained within just the hours of school, at least, in the past. Now it’s 24/7. You’re putting on your own Reality TV Show.

And you’re essentially becoming an agent of feeding what the algorithms want. “If you do this for me, I will put you into these people’s feeds, and therefore you’re going to have a higher chance of getting likes. And you’ll get a bunch of dopamine from it. And I’ll know that people like you, and then I will start to fit this into their news feeds.” And these systems of people basing their actions on what the algorithm wants.

And this is something that the author Yuval Noah Harari–he wrote “Sapiens“–everyone a couple of years ago read that. His follow up was “Homo Deus,” and he basically talks about the algorithm as God.

People are like, “I bow down and serve.” So I think framing technology as a tool for you to use versus the other way around is the overall mindset to take to make sure that you end up on the right side of this helping you do things more effectively, or efficiently, or that weren’t possible before. And let you accomplish amazing things.

And the other side of it is adopting a mindset of always learning. Because that’s what advantages that technology has is that it’s always improving and compounding. I was more speaking in terms of technology versus us. What advantage does it have? Humans are always making it better.

Whereas by default humans don’t necessarily… They’re not on some sort of optimization curve where like by default we’re just more fit and strong and smarter. We have to actively go for it.

And so using a tool with a learning mindset and being competent, understanding, self-aware about what is my circle of competence today? It’s a concept from Charlie Munger–Berkshire Hathaway. What am I actually aware of and have a working mental model of and can act on today? And what is just outside of that circle of competence? And what are the steps I could take to get there?

And it turns out usually technology makes it easier. Is one of the things that’s the obvious thing to let you expand that?

Just looking at the sheer amount of information available on the Internet, there’s probably some kind of self-guided learning you can do to accomplish parts of your ends. Or to figure out what the path is that you should then be pursuing. I think it’s kind of the combination of these two mindsets–assuming that you always have to be learning, and trying to use technology as a tool versus the other way around.

Mark: So to take a hard look at how you use technology, and to recognize when technology is distracting or pulling you away. Or keeping you in your circle of concern or knowledge–competence right?

And then where you have opportunity to use technology to expand that circle. And to gain new competence. I think that’s cool. Most people have trouble going into their discomfort zone.

Kyle: Right.

Mark: So that’s where I come in is to help people get comfortable with discomfort.

Kyle: It’s really hard to acknowledge especially your weak spots. Where am I deficient? You want to focus on the parts where you’re good.

Mark: So we need an app for that, Kyle. That’s your next project.

Kyle: (laughing) What’s a start-upy name for self-awareness? Self-awareness but no vowels. “Slf-awrnss.”

Mark: (laughing) Build it on the blockchain with AI on top.

AI and Consumer Prices

41:13

Kyle: But I actually genuinely think that something like that is actually going to exist. So this is a theory about where is AI going to have the least expected impact? You look at what AI is accomplishing today. It’s recognizing specific classes of objects in images or video way better than ever before. And it understands human language way better than ever before. And it can act in the world way better than before, because it can see in every direction, and build a 3D map and just know all the stuff not to hit or trip over or run into.

But most people especially like an event for people who are trying to think to how these things will fit into their business. Their world. Those are really abstract and kind of hard to connect.

At a16z one of the theories that we had about where AI would have this surprising, deep actually kind of paradigm changing impact would be the areas where if you look over the last couple of decades, costs, expenses have accumulated far faster than the rest of the economy. So if you look at things like consumer goods, clothing, electronics–you can buy a 65 inch TV that’s 4k and beautiful for $1000. Where 10 years ago it would have been 5 times as thick and 10 grand. Consumer electronics always becoming cheaper and more capable.

Clothing similarly, on this curve of becoming much cheaper.

But if you look at things like education, healthcare, construction–there’s a reason that students are in ever more debt. There’s just… inefficiencies are actually accumulating in these markets. And one of the things that we basically talked about as a theory would be if you were just coming out of machine learning PhD program at some top school, and you wanted to solve a big problem, but had no industry experience–because, again, you’ve been in academia for 10 years or whatever it is.

Pick one of those markets, find the big categories of companies in those areas. Figure out the rough connections between them and their workflows. Identify friction. That’s where you go. That’s your problem to solve.

Mark: Sounds to me like Singularity University. Peter Diamandis‘ mission. You come in here, and you come up with an idea that can impact a billion people. And then we’ll help you execute on that idea. That’s pretty cool.

Kyle: I tend to be of a mindset of solve a very specific pain-point and work your way up to that billion person impact. But overall, the mindset of there are just opportunities that are ripe for this kind of disruption…

Mark: But we have this crazy period right now where everything can be disrupted theoretically. We have pretty much within 5 years or so, everybody wants to be connected and have the ability to be connected. Meaning they’re not trekking for water or food somewhere. Will be connected ubiquitously with Wi-Fi. Probably a free device provided by your VC firm.

And hungry… not for food, but to participate in the global economy. And to solve problems. So I think it’s like an extraordinary time for an unbelievable amount of really good things happening.

Like, people really wanting to solve the environmental problems. Go to space. Mine Mars. Set up colonies. All sorts of crazy, cool things happening.

Kyle: It turns out there’s like one guy making people reimagine those possibilities…

Mark: Right. Inspiring people. Yeah, yeah. Elon’s incredible.

But thanks for being part of that, by the way. I mean, you’re 24 or 25 you said?

Kyle: Couple weeks I’ll be 25.

Mark: Yeah. So it’s those of us who are in the Baby Boomer… we’re enjoying the ride. And some of us get to play as investors and everything. But you were brought up with the Internet. I was brought up on my friend’s farm. You know what I mean? I was poking holes in the earth with a sledgehammer and baling hay when I was your age.

Well, actually, I was in the SEAL teams… Little bit different.

Kyle: Yeah. I mean…

Mark: Isn’t it exciting? Do you get excited?

Kyle: It is. Every day I go to a place where we are making one of the most advanced robots in the world. And it’s also something that any consumer can just buy online. Which is wild.

And when you talk about these big things… Opening up space travel to actually be a possibility. Or significantly revamping the infrastructure around cars and how we power our homes and commercial buildings, things like that.

For a lot of those areas… robotics and autonomy gets a lot of this too, where there’s kind of obvious downsides coming down the pike, and people would prefer… some people would prefer that, “Let’s just not even go down this route. Let’s not replace the folks working places like Fast Food Joints. Those jobs, we’ve had them for decades, parts of our economy have kind of warped themselves around their existence. Let’s leave the status quo where it is.”

Mark: You have to reimagine the whole economy.

Kyle: And this is where…

Mark: There was the Industrial Age economy where you get a job, pretty much if you want a job. Except for 6 or 7% of the population. And then you’ll go get it. Then we social safety nets.

That’s just a concept… Just like what Harari said in “Sapiens,” that’s a myth. That’s a story. And we all buy into that story.

But guess what? That story isn’t going to hold up in 20 years, you know?

Kyle: And the thing is, it’s much easier to identify the things from today that will change. But it’s so hard to understand what will arise because of these kind of assumptions and the underlying economy changing.

Mark: The story of what it means to be human is being changed by our creation. Which is technology.

Kyle: I mentioned this earlier in our conversation before we started recording but Marshall McLuhan, the philosopher who… he’s most famous for “the medium is the message,” line.

But he talked about new technology as both extending humanity as well as cutting elements of it off. For example, the telephone gave us the ability in real-time, communicate with someone on the other side of the country and the planet. But it also killed penmanship.

And that was something that for a couple hundred years was part of being a cultured human. “Oh, my handwriting is very good.”

Not just that you write well in terms of your prose, but actually like the actual art of it.

And it’s from little things like that, all the way up to how do most humans spend their time?

On the other side of it, we’re probably going to end up feeling pretty happy about where we land. The idea that automation will kill all jobs, and people will be just on the dole, I find highly unrealistic…

Mark: They’ll find other ways to find meaning. And to add value.

Kyle: People aren’t going to stop looking for those things. Even if… let’s say there’s some tiny percent of the population that is, “You know what? I’m going to take a paycheck from the government or from some blockchain every month and play video games or be in VR all the time.

One is that the worst thing in the world? Compared to how young men who have been down on their luck, haven’t found opportunities have spent their time in the past? Debatable.

Mark: But there’s whole economies in those video games.

Kyle: That’s the follow-up point which is we’re moving to this point where digital goods matter as much physical goods in some areas. And…

Mark: Yeah. “Ready Player One” brought that home…

Kyle: Right. And the idea that… and this is something that has existed for more than a decade now, there is a video game from the early 2000s called “Second Life.” So people would make outfits for virtual characters, or cool apartments or activities that you do with other people.

Mark: Yeah, there’s hundreds of thousands of dollars of economy going on…

Kyle: And they make six figure incomes just from making virtual things to do.

Mark: And there’s factories in China where they’re just playing video games to build avatars and then they’ll sell the avatars who have all this value. I don’t know much about that, but that’s just insane…

Kyle: And so you get to this point too, where you think about… so let’s say we have augmented reality where we’re able to while out in the real world, you’re essentially having virtual goods overlaid on top of things. And maybe it’s like when people see me through their AR glasses, they see I’m wearing a nicer outfit than I am actually.

Maybe it’s things like that. Maybe it’s kind of small, widgety apps that pop up in the appropriate context. You’re walking down the street, and you want to know just “where are the good Thai places?” And it just “bloop-bloop,” overlaid on top of your vision.

All of those things you could basically imagine creating those virtual goods. It allows entrepreneurship with far lower friction than was possible. If you wanted to make sunglasses for the masses in the era of physical goods, you had to spin-up production somewhere…

Mark: Now you can just buy a 3D printer and take some contracts…

Kyle: Well, no. I was going to say in the era of digital goods, they’re bits. You literally enjoy control. And that’s something that’s something that’s kind of another lens I used to look through the world at when I was an investor at Andreessen Horowitz. Was if you look at where technology ends up being really transformative is if you look at the world as a series of equations. Representing like, friction or costs or how people value things–what really impactful tech does is it makes numbers in those equations go to zero.

So what did the collective action of Facebook, YouTube and Twitter do to media? It made distribution of information–the cost of it zero. And that’s why newspapers went away is because they were based on a world where part of their model was distributing information had physical constraints. You had to print papers and then distribute them to people in a certain radius of wherever you made them. And so newspapers enjoyed this geographic monopoly.

And the same thing for local TV stations. And by like, virtualizing the ability to distribute things we mad the cost of that zero. And radical changes happened because of it. And so that’s again why I had that model earlier of look at these areas where there’s always increasing inefficiencies. Making any of those things go to zero, will entirely change the world.

If you can make construction… you know that no subcontractor will ever mess up any part of the job making a large skyscraper. That can add billions of dollars to a project, and it makes… decreases the amount of investment in those kinds of things. And if you eliminate that entirely? How much more dense will urban areas get over a 100 year time horizon?

Mark: Right. Interesting.

Or if we’re having virtual meetings, and we don’t need skyscrapers anymore…

Kyle: Yeah, so that’s another thing where the collision of these different areas also introduces radical transformation that’s hard to predict.

So, for instance, virtual and augmented reality makes face-to-face interactions possible without actually being in the same location. And so…

Mark: But then it cuts off human contact.

Kyle: Well, unless again, it’s taking in enough about the world so that you actually feel like you’re getting all of the body language that people give off as they’re having a conversation. And you’re going to have gloves and jackets that let you actually like, hug and have handshakes that pretty accurately represent what would have happened in real-life.

You mentioned climate change and all this tech being affected by it. It might not be installing solar panels and batteries everywhere. And all of us switching from internal combustion engines to electric vehicles anytime soon.

It might be that virtual and augmented reality obviate the need to do business travel for some percentage of trips that happen every year. And if you look at an average person’s impact environmentally in terms of carbon output, you can switch to a Prius or an electric car versus your CRV. Doesn’t have much of an impact.

Cutting off all flying–far more impactful. An order of magnitude more impactful.

And so it might be, like, entirely unexpected technology changing how we go about our lives. Changing those underlying assumptions. Changes our underlying behavior. And that’s actually how climate change–the accumulation of carbon dioxide in our atmosphere–that actually might be what moves the needle on that.

Mark: Wow. Well I think that’s probably a good place to kind of surrender…

Kyle: I could go for days…

Mark: Because we could go for days. This has been fascinating. Kyle, thank you so much for your time.

Kyle: My pleasure.

Mark: You rock. So is there any place that people can learn more about what type of thoughts you have? Do you have a blog…?

Kyle: Yeah, so I’m on Twitter pretty heavily. I made a lot of jokes about social media addiction. Twitter’s where I’m addicted. So my handle is @kylebrussel.

And from there you can also find kind of longer form thoughts that I put out on Medium. That’s where I put kind of articles that I pontificate about some of these things…

Mark: Like you did today. Well thanks so much for your time. And I look forward to reading your stuff and learning more. It’s been a lot of fun.

All right, folks. That was an interesting journey… look into the future. It’s all coming at us really fast. So stay focused and make good choices. Be unbeatable. Train hard. Stay focused about also keeping your humanity.

And not surrounding or outsourcing too much to technology. Use it as a tool. Right?

Kyle: Yup.

Mark: Hooyah. See you next time.

Leave a Reply