From left, David Kirkpatrick, Tim O'Reilly, Max Levchin. (All photos by Asa Mathat)
Tim O'Reilly (l) and Max Levchin.
Tim O'Reilly (l) and Max Levchin.
From left, David Kirkpatrick, Tim O'Reilly, Max Levchin.
President and CEO, HVF LLC
Founder and CEO, O'Reilly Media, Inc.
Founder and CEO, Techonomy Media
As tech- and connectivity-driven change floods across business, social life, government and governing, international relations, jobs, education, health care, and any other human activity you can name, where are we heading? How different is the world we are entering, what must we do to mentally and practically prepare, and what should we resist?
Read the full transcript below. (Transcript by Realtime Transcriptions.)
Kirkpatrick: So to wrap up we’re going to try to have as big a picture conversation as we can. Come out, guys. I think we have the right people to do that.
Tim O’Reilly, who is really one of the great thinkers and also conveners of technology, widely credited with having inventing the term “Web 2.0″ and then helped run a conference with that title, and has been in the business of publishing information and books about technology, both at the sort of the functional level and at a higher level for long time, and kind of innovated in the conference business in a number of ways. Among other things, he now runs a data conference called Strata that’s really trying to conduct cutting edge-conversations about big-data analysis and what that all means.
Max Levchin, one of the renowned PayPal mafia. He was CTO and cofounder of PayPal, and the people he worked with are Elon Musk, Reid Hoffman, people who really have, like him, had huge ongoing impact in the overall space and world in which we live.
So at the moment Max has his own sort of personal incubator called HVF, yes. And one of its most prominent current outputs is a company called Glow, which is intending to help women get pregnant, basically, help use data to improve the process of fertility for ordinary people. It’s very much in aiming at an “access for everybody” kind of mindset.
So I want to start by asking both of you, what is the biggest thing we should be thinking about in order to understand the future? And I want to start with you, Tim. It could be more than one thing.
O’Reilly: Yeah, it’s very hard for me to ever boil things down to one thing.
The first thing I would say is I always look for areas that are pre-commercial where people are having fun. So, for example, when we launched Maker Faire and Make magazine in 2005, nobody thought there was a lot of money to be made in things like 3D printing or all the maker stuff that’s now the rage.
So I ask myself, where are people having fun today? I think you’ve already had them on stage. It’s areas like biotech. BioCurious. We have a bio newsletter called BioCoder. That seems to be the area where people are having that pre-commercial, “Oh gosh, this is so cool.” So that’s one thing I would say.
The second thing is that it’s really important when we think about the future to think of trend lines and probabilities. And probabilities involve a range of outcomes and the likelihood of those outcomes.
So I always get a little fed up when I hear people go on about the straight line to the singularity, you know. And I think, “Oh, my gosh, I was a classicist and I remember that Rome fell, and, for example, the Hagia Sophia in Istanbul was the largest building in the world, for a thousand years. And the next time somebody built something that big was the Blue Mosque half a mile away, which is an exact copy of it.”
The technology stagnated for a thousand years. Then it came back, and it came roaring back, and that’s fantastic. But I certainly see in our present lots of things to be scared about that, you know, could really derail our assumptions of progress.
Kirkpatrick: Well, that’s interesting.
Before we get to my question to you on what those are, Max, what would you say in answer to what should we most be thinking about in order to understand the future?
Levchin: Definitely very hard to cut it down to one or even two things. But I think the—it’s very hard to think about medium-term future. Short-term future is pretty easy and very long-term future is pretty easy. We’re going to live forever, we’re going to have computers embedded in our heads, and everything is going to be pretty awesome.
Kirkpatrick: You mean that seriously.
Kirkpatrick: You really think we’ll live forever?
Levchin: Some definition of forever. There is really no good reason why what we have inside us cannot be replaceable and repairable, because everything outside us has basically been made repairable or arbitrarily—we can predict when something will break upon a million use, a thousand use, whatever it is. We have that pretty well figured out for kidneys and hearts and major organs.
There is no reason why we can’t start planning. Capacity planning is something we’ve gotten really good at in IT space. There’s no reason why we can’t do capacity planning in the human body.
O’Reilly: I want to pick up on that when you’re done.
Kirkpatrick: Back to medium term.
Levchin: The medium actually is really hard. I’ll tell you the pieces that fascinate me, personally, as opposed to the global landscape.
I think many times you sort of see things that are just going to just continuously change, and that’s what I look for. So I love the idea of looking for pre-commercial fun, because that ultimately spells out a huge commercial opportunity.
The thing that I look for maybe slightly shorter term, and that is something that looks like it’s bubbling up and trying to make the prediction of what does it go from bubbling up to just boiling over.
Kirkpatrick: Where do you see that right now?
Levchin: I think there’s a lot of that in biotech. I think genomics is one major breakthrough away, which is kind of amazing. I think the idea of human-computer interfaces at large is basically tied to one or two or three breakthroughs in brain science.
Kirkpatrick: So Google Glass is not where to look, it’s brain science is where to look for that?
Levchin: I think Google Glass and several other human-computer interface hacks where the hardware is visible and is unwieldy—even Google Glass, as cool as it is, is still pretty unwieldy. It wouldn’t work for me very well, since I have glasses that I have to wear. Once we get out of having to create large lumpy hardware to interface with our heads, inside our heads, and go directly to it, I think it’s going to be a huge change.
I think the general idea of healthcare changing from corrective to preventive model fundamentally is what interests me the most. That’s where Glow came from, the idea that you need to do capacity planning on your body. The only way to do capacity planning is you monitor usage, build models, and forecast breaking points.
If you start tracking people and their health signals, primarily for themselves, you can do much better forecasting and therefore price reserve or correct before a real danger strikes. From that I think will come the ability to predict much more drastic changes such as deadly diseases or even accidents, and that will eventually lead to more or less forever living.
Kirkpatrick: You had a question for him?
O’Reilly: Well, more— yes, I guess it’s a serious question. You made the comment that in electronics and in our devices, we’re getting more and more a sense of how to fix things, where they break. And yet as a culture, what we have chosen to do is to make those devices more disposable, not last forever. And why do you think it will be different with people?
To me one of the real risks is, yes, we get this technology of life extension, and it’s reserved for a very few, very rich people, and everybody else becomes more disposable.
Levchin: I think that’s a really dystopian view of the world. I’ve seen movies to that effect.
Actually, I think we will go more towards disposables. I quibble with the idea of disposable devices. I see it as interchangeable and replaceable devices. We eventually go to building blocks. “The Mythical Man Month,” one of my favorite computer science-related texts describes the process of building an operating system in a year as a monolith. The people reading say, “What the hell were they thinking?” Of course you break it up into lots of parts, you test them separately, you mash them together, and you might miss the deadline by a couple days.
The same thing happened across every industry. We build building blocks, we make sure the interfaces are standard, we combine them, we replace the part that breaks. The rest of the system works, but the part that broke is disposable.
I think that’s what’s going to happen to the human body. We will look at the core value that is the brain and whatever passage for soul these days needs to be preserved forever, but you have the capacity to regrow a nail as you cut them off. You should have the capacity to regrow a finger if that somehow gets damaged. We should get to the ability to regrow or reimplant an eye or—we have it for hair.
O’Reilly: Totally buy that, totally love it.
I do think, and this kind of comes back to my point about the things that could derail that trend: religious fundamentalism, anti-science, political instability, sort of lack of political will to keep funding certain kinds of things, and a sort of breakdown of economies, you know, wars that bankrupt the world. There are a lot of things that could derail our happy consumer utopia.
I think it’s more fragile than it appears, and I think we need to take seriously the job of building our society as a fair, just, open society; rather than one that is, I think, in a lot of ways becoming less good than it was all the while that it could be getting way better.
Kirkpatrick: This came up in the World Economic Forum breakfast at our table quite powerfully, the same general point. The key question there so often comes back to, we are so untrusting of government’s ability to take a grip on that kind of question. And it sorts of leaves us with big companies.
And then the question is, can big companies begin to start to have a sufficiently enlightened view of their own and our collective self-interest that they can play a role in helping move toward that future?
Or is that just inconceivable to you?
O’Reilly: No, it’s not at all inconceivable. I think already we have a lot of big companies that are in fact shaping our future. It’s a really useful lens, for example, to look at the world through and see how, for example, foreign policy is no longer just between countries; but between companies and countries.
For example, Google had a foreign policy with regard to the Egyptian revolution. Big financial firms, hedge funds, had a foreign policy with regard to the fate of the Euro a few years back. In fact, I like to frame what was going on back then as a cyber war between hedge funds and the European Central Bank.
Yeah, there’s a lot that corporations are doing to reshape the way our world works. I think some of them are doing really good work, and some of them are actually fairly inimical to positive social values.
Kirkpatrick: But the general direction in terms of corporate values? Let’s start with the United States. How would you describe that?
O’Reilly: I would say in general there’s this huge acceptance that it’s okay for companies to screw their customers.
O’Reilly: Yes, I think it’s very—I think in Silicon Valley there’s certainly much less of that. But if you look at airlines, for example, the whole model of sort of hidden fees and various kinds of user-hostile behavior and policies. Banks, the big banks, literally raping the economy in order to fatten their profits. You know, this is a terribly dysfunctional system where companies no longer feel they have a social responsibility.
Kirkpatrick: Max, do you agree with that? And if you do, do you see any mechanism that could counterbalance that, help us move more toward the enlightened view we mentioned?
Levchin: I think the problems described are real. My view is more optimistic in the sense that I think we hit the rock bottom of short-term thinking, which I think is at the root of all these problems, where people basically look at their company as a vehicle for personal enrichment, be it a huge company or a small company. Once the interest of the actual actors goes all the way down to the personal wallet, it’s very easy to see how they have a completely myopic view of the long-term impact or even broad short-term impact.
The changes that are happening, at least on my viewable territory in Silicon Valley, is that guys and girls coming out of college are no longer interested in going to a place that has the best view to go public soon and move them into a better car. They’re actually, you know, every time—so I run this company called the Firm in addition to Glow, which I don’t run, I just chair. And the Firm is an attempt to sort of reimagine what a modern bank, modern credit institution looks like. It’s all about transparency, it’s all about clarity, it’s all about speaking directly to the consumer and being honest. And all the people I recruit in the company—and I interview every single developer we bring on—it’s remarkable how these 21-, 22-year-old guys, fresh out of school, basically say, “Look, I want to work for a company that has real aspirations, that wants to change the world, make it a better place. I want to work for a company that I’m proud to tell my parents. And by the way, the reason I’m here is because my parents really took it on the knees during the 2008 financial crisis, and I love the idea that I’m working on something that is a direct counterpoint to the irresponsibility, lack of transparency, density and B.S. of the too-big-to-fail universe. I want to be on the other side of that.”
O’Reilly: I would totally agree. This is a huge upwelling of idealism, and the startup community is a super—can be a super-powerful force for good. We certainly see that same thing with Code for America.
Kirkpatrick: Which is an organization that you are deeply involved with.
O’Reilly: Yes, I’m on the board and very deeply involved with.
People come for a year for a stipend of $35,000 to come work and do tech work with cities. And they are leaving jobs at Apple and Google and places like that. That’s not people who are like, “Wow, I want to get rich.”
And then they go on and they’re starting civic startups to try to actually make government work better, because they realize just how critical it is that we have public services that actually work, and that we don’t sort of just accept that, well, too bad for the poor.
Levchin: One thing that’s really important in these sort of Code for America type organizations and just the general thing that actually is very exciting to me, is for a long time, charity and good works and giving back and civic society and all that stuff is not new. It’s been narrative of the American view of the world for a very long time.
Until about three or four years ago, one of the prevailing themes that really just rubbed me the absolutely wrong way was part of the story, this ridiculous “do the best you can and that’s good enough,” which is complete insanity. It’s not good enough. It’s good enough when you start improving things. Good enough when you go to sleep feeling like you’ve done everything you needed to do.
You haven’t benefited anybody, but you’re okay? That’s the same thing as lining your pockets at someone else’s expense, basically. You’re just lining your conscience at someone’s else’s expense.
So the thing that’s really changed, I seek leverage. When I go and I put my hard-earned dollars to a charity, I want to know that 99.9 cents goes to benefit my country, my people, my city, my county, whatever. If I’m volunteering my time, if I’m Teaching for America, Coding for America, or becoming involved in a civic startup, I don’t want to give money to one person; I want to give effort and money and whatever else, and benefit as many people as humanly possible. I actually want to correct the path that needs correcting. I think that’s a really great trend. That’s the thing that’s most exciting.
Kirkpatrick: You know, both of you are data-centric thinkers. You run Strata and you’ve done a lot of other stuff. Glow’s whole idea is to apply it to data, and I’m sure you have it in every other activity you’re doing.
Given what we said before about the need of—I like to call it understanding more clearly our collective self-interest. Do you both believe, or either of you, that data, properly analyzed and displayed, could help us get to an understanding of our collective self interest in order to correct some of the risks?
You gave a very long list of things that could go wrong. Many of them are remediable if we have a collective understanding of what really is in our best interests, right?
O’Reilly: I would say many of the things I’m worried about come from people who ignore the data. So I’m not sure that the data is the solution in that sense.
I do think, though, that we do need a revolution in the way that we use data, for example, in delivering, you know, public services, and in regulating public services.
You know, I had a conversation once—this was back when we were fighting over SOPA and PIPA with Nancy Pelosi, and she said, “We have to balance the interests of the tech industry against the interests of Hollywood.”
My response was, “No, you don’t. You have to figure out what is right for the citizens of the United States. You may want to give the—whatever to both sides. You know.”
How do you do that with data? You measure it, you figure out what works. And there is something very powerful in the way that startups are working today, with the lean startup methodology, of identifying a plausible problem, building something that you think may address that problem, and then really measuring.
I contrast this with a conversation that I had with someone in government. I said, “Do you actually measure, like, how many people use a website—this is long before the Healthcare.gov debacle—use this website that you’re spending tens of millions of dollars on?
And their response was, “That would be a good idea!”
It’s like, “Oh, my God. Nobody is even measuring this stuff!”
Kirkpatrick: So this is my point. There are plenty of people who just don’t care about data. We have an anti-science subset of our culture. But there are plenty of people who probably could and would use it that maybe would make a difference.
I’m just curious of your thinking, Max. Can we get to a better bigger-picture view of our future that will help guide us?
Maybe even an entrepreneurial opportunity to help do that, I don’t know.
Levchin: I think so. I’m obviously a big data nerd and data junkie, so anything we can do to make the world a better place with data is exactly my cup of tea.
I believe in thinking in terms of products. One of the sort of thought experiments that I thought would be really fun is to have a giant infographic on some large display in Times Square or every large display in America at noon or 1:00 every day. Just teach the citizens of whatever locale, here is what is going on, here is the most important stat of the day.
Most people skip over the stats that are published on the front pages or back pages of their local newspaper. Just tell them on a daily basis, here is what happens to your dollar. Here is how your taxes get spent locally. That would go a long way.
Kirkpatrick: I started this conference showing the Gapminder.org, health and wealth going up over the last 200 years. That is a fundamental concept. People need to think about these big data realities that are great. That’s great.
O’Reilly: Hey, I want to move a little bit away from data. You were asking about startups and startup opportunities, and we touched very briefly on Google Glass. But there’s a really big idea that I think startups need to be engaging with, which is, as technology gets beyond the screen, or even when it’s on the small portable screen, but it gets out into the real world, we have the opportunity to rethink the work flows and the interfaces, not just to the device and the application, but to a real-world process.
Uber is a great example of that. There was a great Tweet from Aaron Levie of Box.com, where he said Uber is the $3.5-billion lesson in imagining the way the world ought to work rather than rebuilding the way it does work.
The fact is, Uber is software above the level of a single device. You have an app, the driver has an app. There is a whole bunch of platform stuff in the background, and all this makes it possible to completely rethink how you get a taxi or get transportation.
It’s just the beginning, I think, of process reengineering for the real world. And I would say when I think of social transformation and how technology is going to affect it, that’s a lot of where it’s going to be. The sharing economy is in that category. There is this huge explosion of things where we’re no longer thinking about technology as how do we build this device or this application. It’s really how do we build this real-world process, which is completely different.
Kirkpatrick: Max, I think you have a strongly held view that in general, tech entrepreneurs aren’t thinking big enough.
Is that a fair statement?
Levchin: It was certainly a very fair statement three or four years ago. I’ve seen enough companies thinking big enough that there’s definitely a trend for the positive. I was very pessimistic about the future of Silicon Valley and American innovation in general four or five years ago. As of about a year ago, I stopped worrying.
Kirkpatrick: Is this the Peter Thiel-ian sort of mindset, who is your colleague and friend? Former colleague.
Levchin: I think both of us have sort of decided that we’ve basically said enough negative things, whether because we were ranting or because the world just changed.
O’Reilly: There are a couple cool things happening.
Kirkpatrick: I want to hear about a couple of cool things that are happening, and then I want to talk about China.
Levchin: I love talking about China. There’s lots of cool stuff that’s happening. I am an investor in a company that’s literally trying to solve every disease. Not just some disease. Every disease. Everything that can be thought of as a deceased cell, they’re going to fix that. That’s pretty damned big.
I’m on the board of Evernote, that describes itself consistently as a 100-year company, which is exactly the sort of thing to do or to think if you are trying to benefit the world, as opposed to your shareholders right now.
Some of the—I’m kind of into biotech right now, so another startup that I’m involved with is testing, in second stage of cancer drug-testing. So trying to cure all lung cancers.
These are really fundamental breakthroughs, and these aren’t like two dudes in a bedroom saying, “Hey, we’re going to go cure lung cancer and let’s see what happens.”
They are pretty far along, which means that they have taken the risks and raised the money, made mistakes and corrected them. So that’s very exciting.
A lot of what’s going on in the world of sensors is just fascinating to me, because ultimately—I think, and Uber is actually a great example. I’ll go back to it in one second. But part of my general investment and concept in product and company development pieces is that we are living at this amazing time of extraordinarily cheap, very plentiful broadband, which can be generally thought of as free.
We can manufacture hardware that is very low power, very wearable, and put it everywhere: on our cars, on ourselves, in ourselves at some point. So basically sensors are just little tiny capturing devices that grab whatever data is available, transfer it through free broadband into the center of the universe—center of computing universe, anyway—where interesting insights can be gotten, and transmit it back to the actors that are providing the data.
If you go back to sort of 30, 40 years ago, the whole idea of yield management was invented by the airlines, and Sabre and all these companies were built around it, where they would gather all the system data from competing companies, centralizing intelligence, and return it back to the participants without compromising their ability to compete with each other, because the data insight gotten from the conglomeration of information was so much more valuable than any one of these companies getting smarter.
That idea has been applied to everything. This whole notion of you’re at the edge, you don’t know much, but you know specifics. Let’s gather all together, even if you’re competing with these people, and give the general insights back to the edge to make everybody smarter and better is very, very powerful.
Uber did that with cars, did this with homes. There’s going to be a company in every imaginable segment doing that approach. It’s only possible because everybody has a phone that is a sensor, the gathers the data, sends it up to the center, and the center has enough computing power.
The dystopian thing, to give Tim some interesting things to argue about, is—not actually dystopian, just an interesting question that I came up with a few days ago. One thing that happens as you have all these sensors, all these watches, laptops, and phones in your pocket, on yourself, we have an incredible number of CPUs with not a lot of memory.
What happens in the world where you just have processors everywhere, that are generally sitting idle, by definition, because low energy requires that they’re not burning too many cycles? What is going to happen to all these chips? Are they all going to run malware?
Kirkpatrick: Maybe they will, yeah.
O’Reilly: I would just second the enthusiasm. We are about to undergo a technology revolution that’s actually bigger than everything that we’ve seen before, on a lot of fronts.
So if we can get the politics and the economics to go with it, I think we really are, you know, heading for some really, really very positive, very interesting times.
But I do think that it’s super important for everybody in the audience and for everybody in the industry to take seriously the idea that we actually have to invest cycles in making sure that our government is up to speed, to the extent that we get—we have this whole world of people who are operating in a world that, you know, the government just isn’t.
You know, take this Healthcare.gov idea, where it costs $600 million to build something that doesn’t work that really could have been built for a tiny, tiny fraction of that by people with a clue. And you go, “Wow, there’s this huge disjunction between the way our big systems work, our governmental systems work, and the way Silicon Valley works, for example.”
Kirkpatrick: Maybe we need some new government contractors to come out of Silicon Valley.
O’Reilly: No, I think we need a new government contracting system. It’s not like there’s no technological capability. It’s the system and the incentives and the way people are selected. I mean, if you have—the guys from development see it, who actually worked on—they build the Recovery.gov site for a fraction of what everybody thought it could be built for and they said that actually without the procurement regulations, it could have been an order of magnitude less than that. They were one of the subcontractors working on Healthcare.gov, and their part worked fine. Because it still costs way more than it should because of the way the system is designed.
A lot of this has to do in some ways with how do we rethink regulation, because it’s actually the federal acquisition regulation that makes things so expensive, because we’re trying to protect against abuses that happened 100 years ago.
Kirkpatrick: Fraud, yeah.
O’Reilly: And various kinds of fraud.
It’s as if Google was trying to protect against click fraud by making this incredible infrastructure of rules that had to be enforced and you had to get approval before you could advertise. It’s like, “Wait, that’s not how it works.”
You do this algorithmic detection in real time of problems. There’s a whole different way of doing regulation, for example, that could completely transform, say, new drug discovery or IT procurement if we just sort of accepted a different way of doing things.
Kirkpatrick: I just want to add a geopolitical overlay, because we still live in a world where you both are very optimistic. You’ve said super optimistic things. But how does that overlay affect what you’ve said? And particularly we heard, you know, that China is roaring ahead in biotech ahead of the United States for a lot of reasons.
I know that Glow is a company that’s more Chinese than American by some definition. So—
Kirkpatrick: Is China a place we should be looking to, to contribute to this progress, for example?
Levchin: I’m actually pretty optimistic even on China, which is a relatively unpopular view. I think one of the greatest fair criticisms of China and kind of the root of the accusation, probably correct accusation, of their general policy of “IP theft is okay” is the 20-odd year history of “We’ve got to catch up.”
China looked at the U.S., at Silicon Valley, the Silicon Valley-like places in the U.S. and Western world and said, “We’re behind. We’ve got to get there as quickly as we can.”
If you look at the copy-cat culture, “Hey, this worked in Silicon Valley; we have to bring it to Beijing.” All that stuff went full steam ahead for the last 10, 15 years, including legal copying and illegal borrowing.
O’Reilly: The U.S. in the 19th century, exactly the same.
Levchin: Right. The founding fathers were, by some definition, criminal. There is a history precedent for everything.
So the interesting thing—I go to China pretty often because Glow is based in Shanghai, although it does build products for the Western world, primarily for the U.S.
Kirkpatrick: Building products for here in Shanghai?
Levchin: Uh-huh. The team is phenomenal, and they happen to be in Shanghai. It’s an international, global world.
The point I’m trying to make about China is that I have a relatively long view of how China’s internal entrepreneurial culture has developed. There’s a lot of new ideas in the last 12, 18 months that are coming out of China that have no precedent in any other part of the world.
China is now actively originating concept. They are creating IP, they’re creating business processes, they are creating business models. Which means that in another number of months or years, they will flip the fundamental view of intellectual property. They will start to want to protect their own, because they are generating plenty. That’s what typically happens with countries that go from catching up to originating.
So that gives me a lot of positive—they will eventually have to come to the economic roundtable and say, “Look, we’ll trade you IP for IP, we’re trade you resources for resources.”
One of the other sort of economic changes that I think is really happening, China is losing its spot as the cheapest place to manufacture and get labor. It’s now rapidly going to places like Mexico and India, which is interesting, because the whole globalization engine brought China up and now places them pretty close to U.S. and other places.
So, once again, they are going to look for cheap manufacturing. All of these things are positive transformations. They are going to normalize international relations, not make them more drastic. China will have a very fundamental economic disincentive to be a bad actor in the cyber warfare or cyber spying, cyber theft.
Kirkpatrick: Any thoughts on that, Tim, whether your view that we’re at the threshold of this incredible innovative age, is that something that’s going to flow from a global set of sources?
O’Reilly: I guess all I would say is that anybody who wants to think about the future should always have very large in your consciousness that the future may not look like the past in some really fundamental ways, and be prepared to be surprised.
Kirkpatrick: That’s a wonderful way to end, which unfortunately, because everybody’s got planes to catch, we have to do.
I really appreciate both of you being here with us to finish on a very large-picture approach. You’ve done a great job doing it. Thank you.