Ladies. And gentlemen please, welcome Alan, blue. Good. Morning everybody. My, name is Alan blue I'm one. Of the co-founders here and I'm also work. On the product vice, president of product here and. Thank. You all for joining us for this edition of the LinkedIn speaker series we. Put these on on a regular basis, to. Try to bring some of the smartest and, most thoughtful, people here to LinkedIn people, are gonna be able to talk about the stuff that matters to us as a company but also matters to the vision, and mission that we have in the world and, for. Those of you watching online and even those here if you, if you want to go to speakers. LinkedIn, com, you. Can check, all the old ones and see what's coming up on our speaker series so, as you all know the, question. Around the future of work is a big, deal for us here, at LinkedIn, not. Only do we concern, ourselves with where we are today but, because we are members, first because we want to make. Each individual. Successful, to provide each individual, with, economic, opportunity, we also want to think about the future of work, is and it. The future is, both easier to see and harder, to see that it ever has been technology. Is changing things very, very rapidly, and. There. Are beginning to emerge a set of voices, around. The world who are beginning to think about these topics, and we. Are very pleased today to have two of those voices two of the people who had originally, started talking about what, it means when, machines. Begin, to, act and. They'll talk more about it begin to act not just in, lieu, of our muscles but also in lieu of our brains and. What does that mean for us I'm gonna go forward basis so, it's the author of two books both, from MIT, and, people. Who have had the pleasure of working with and speaking with on, many occasions. Eric. Ren Wilson and Andrew McAfee please, come on up and. My. Co-founder Reed Hoffman is going, to engage them in conversation, enjoy. So. The fortunate, thing is we're at least a few years off from, machines. Replacing, the our ability, to use panel conversations. So. In terms of brains. So just. So everyone knows I've been a fan, of Andy, and Eric's for a long time I think their work is really important, if, you actually haven't read it I recommend it, and. Just so you all know our main strategy, is to talk to read whenever we can and in secret transcribing. And says this. Secret. Click. Covers on it and try to sell copies plastics. And, so. Let's, start with so. Why did you write this book what's. Your target what are you trying, to do as we know most of the LinkedIn audience knows. The three transformations that are happening but, why, did you write this book. Well. I think we're in the early stages of a huge, revolution.
Andy And I have been looking at the way technology. Has been transforming, so much the economy and all of you in this room have, been working to lead a lot of this transformation but. What you, may not be as obvious to, you is the way the rest of the world the west of the economy, is being changed and we're in a business school and looking at the companies that are being disrupted that are being transformed, and we look ahead and we see the, next 10 years is potentially, being the the, best 10 years in human history or some of the worst 10 years because there's just some huge challenges and the, the gap that we seek to address is, not, that you know the technology isn't moving fast enough you guys are doing a great job on, that it's that the rest of us aren't keeping up with it that the, economics, is way behind the human institutions, the organizations. And so in our previous book second Machine Age we talked a lot about the. Disruption that's coming the bounty that's being created but also the inequality. That's being created and this one we're looking at these three great, rebalancing. These three great trends, reshaping --zz I think Reed says what would be a better word than rebalancing. Between. Mind, and machine, product. And platform, and the core and the crowd and this triple, revolution. Is going. To change the way a lot of businesses need to be run the, way a lot of people Reese, kill themselves. And the, way we think about its society, more broadly after. Eric and I wrote the second Machine Age we went on the road we talked about it a lot and we kept on noticing that we're having the same hallway conversation. Over and over which, is you'd golf stage and you we'd. Be accosted in the hallway by somebody saying essentially, I run. A, medium-sized. - big company, in industry, X I believe. The story that you're telling what. Do I do now. How do I think about the world how do I think about my business my industry, differently, it was a great question and Eric, and I are both Business School guys and we kind of thought it's incumbent on us this is what we're supposed to do to try to provide answers to that question the, problem was we didn't know the answer to that question at that time which. Was a little bit scary but also a great opportunity because, the two of us get to go into a room with a whiteboard and try to figure stuff out talk, to some of our favorite a geek friends out there in the world and honestly try to answer this question actually that's one of the reasons why we're here we're literally looking forward to the Q&A and the discussion, from from reading but from all of you so we can hear more questions about things that we should be thinking about and getting to work on it so another one of our our secret, formulas, to, try, to do better work is to hear from some of the best and the brightest about what you're seeing and when we went out into the world we knew that we had, talked about some number, of things that were happening and one is not enough and I, got really good advice earlier in my career from somebody who said if you ever find yourself on stage saying, and sixthly. You're. In a lot of trouble right, so six is too many it has to be like a finite, number of things that people can get their minds around and that's, where this idea of three three, rebalancing, three reshaping, came, from the good news is that as. Eric and I kept talking to each other we, learned that there are three, separate, bodies of Nobel prize-winning economic. Work that have a huge amount to say about. The stuff that's happening today so the economic, backbone in, the book is one, of the things that I'm proudest, of so. I. Think. I'm a word pet and today it's, not a secret when you announced the web but, um in terms of strategy. Was. It, was a thing. Yes. Exactly so, what are some of the things that there's. Obviously as, as traditional, industries and traditional companies look at the, transformation, that's happening with technology. They have this kind of massive. Li quick changing, of within. The world of software you know artificial intelligence platforms.
The Deployment of a crowd. They. Have a, need. For a different skill. Basis a need for a different strategy. For the business at different decisioning. Different approach to data there's. Obviously lots of things to do and don't write so like one of the pieces of advice is not well, everyone, should go try to become an AI first company next. Year because that would be a mistake, even though AI is going to transform all industries so what are some of the kind. Of the or. The cheat sheet on the do's and the don'ts for. Kind. Of the traditional industry, folks well, let me dive in with one of them which is they should be a lot more data-driven a lot more you know not everyone's going to become you, know LinkedIn or Google or whatever but, if you in terms of resetting, the dial our experience. We looked at a case after case is most companies most organizations, are not, nearly, data intensive enough not merely, technology. Intensive enough and too many decisions are being made by human judgment and we spent time talking to what we mentioned, Nobel Prizes people like Danny Kahneman and we're just astonished. By the number. Of human foibles and mistakes and judgment and decision-making that happen over and over they're systematically there, are about. 175. Entries, and we're gonna run through all of them right and 170. 56 but in. The Wikipedia, entry for human, cognitive bias and so people when the first one is but one of the first ones is people are way overconfident. On their own judgement, so we look at you know just raise your hand if you're a below average judge, I'm. Serious. You're. Foundered right, but. But, but. You know just looking at our owners so we did a bunch of studies you know ourselves and looked at how if you did more data-driven decision-making, you could do a lot better and I just very briefly you know pointed, our own decision. Making in the mistakes we made you know we've tenure, people and we promote people based, on this committee that sits around and makes our human judgments, and along. With some grad, students, and some people operations Research Center we did a little analysis, we could call it Moneyball for professors where if you just took a bunch of data about you know citations. And publications, and other kinds of data and you did some more sophisticated, analytics. On those and. You. Could, predict, who, should, be getting tenure who didn't you compare the list that the model created with the midst of people who actually got created they, overlapped, about 7075. Percent of time so pretty good comparing, it but, what's interesting is that the ones that it disagreed, on the. Model, those people, did substantially, better in terms of having high impact, research their research got, more pub more citations, had, more impact gotten, to higher things in the subsequent years so, if, you wanted to have tenure.
People Who did great research, you would have been better off just, going with the model now our book, is a little more nuanced than that we don't say just throw out the human decision-making but, this was actually a debate that Eric and I had because I'm kind of of the school you should throughout, the human decision-making, eric, has a little more fondness for our wet wear than I do I think you have an algorithm to make your point I kept. On citing actual evidence threes like no screw oh man I'm not judging you separate even generated, a model that showed that I might have been convinced but but yeah this was well that's part of the fun of writing a paper about writing the books and papers with with Andy is that we had these creative. Tension and debates and I remember these discussions, we had over lunch we sat down and and I, think we came to sort of a meeting, of the minds that, as. I said the opening using. A lot more data and, models. Would be very, very beneficial I still, don't think that we can just go all the way that way for most problems but many of them we way better off turning the dial lot further than that direction Reid I think the single most common failure mode that I see among, enterprises. Around, the world that we've worked with is the, managers, the leaders of those companies still think a big part of their job is to be the. Gatekeeper. Of ideas, to pass judgment, on the, ideas, that might make it out into the world or not and they do that based on their gut their experience, their expertise, their what not our. Favorite business acronym is hippo which. Stands for highest paid, person's. Opinion and. It's, how most decisions get made in most places and when we go around and talk to super geeky companies the thing that I think I'm most impressed by is how, the people, that run those companies are trying to get out of the hippo business, and trying, to create a culture that, that lets the evidence speak that has an honest conversation but, what are we going to do and tries, to be a lot less hippo, driven, when. I was doing some of the. Conversations. Behind the, book that I think I'll be publishing this year blitzscaling, you guys have seen early. Drafts of it actually. On this thing one of the interesting, things I discovered was, the. Way that Beezus manages, he says look if it's an opinion it's my opinion. But. If you, have data you, can overwhelm my opinion right, so part of his attempt to kind of make that, drive, is he you have I disagree with you bring, data and so, like one of the more interesting. Examples. Of that was the Amazon. Feature of. Asking. Customers. To write. In almost like Wikipedia, like information about the products bees those thought that was a terrible idea so. What they did is they literally hand, they paper did they emailed. A thousand people created, a little website had, people type in and brought in the we, did a thousand emails this is what we got here's, the lady looks like story yes that's great that's great advice actually, and that's one things we talked a little bit about in the book as well is this, culture, of experimentation, and a/b testing you know scientists. Have been doing this for a long time the experimental, method but businesses, outside of LinkedIn and Silicon Valley it's fairly rare Amazon, to, do those kinds, of tests, but it's so often now because of the digital infrastructure we have now that, you can instead of debating in a room and people you know having my opinion versus your opinion you say well let's run an experiment let's run a test and it could be a strong, like that or if you have a website sometimes you know you guys run I don't know how many hundreds. Of a/b tests each week I my single favorite example, of that that we put in the book is we came out here we interviewed fish to see you of Udacity, and he, said they, were. Employing. People to review code that students submitted as part of their courses and one of the engineers said wait a minute we have alums, out there from all of our courses when, we just contract with them to review other people's code and and, this says I was like hey.
You Know go, go try it and you know go go see if tapping into the crowd in that way will work and what I loved about that was he didn't even attempt, to exercise his own like that seems like a good idea to me kind of that's, rare I mean kids like him and and Jeff, Bezos help you guys are, red. Bull, that's rare spend most of our time nine percent talking to the CEOs the rest of the economy. And it's, a real cultural. Shift, move away from being a hippo to saying okay I'm gonna step back and let the data speak on the in fact that's you know I'll be more biggest, pieces of advice is it's, not so much the technology it's that culture, of how you make decisions that needs to change have, you been thinking about creating, or have, you already created a new course at the Sloan School. Several. Yes. A little bit yes so there's two courses that are well three courses that I've been teaching that that are sort of touching on this so once a ph.d course we get into it all that there's, an MBA course. Economics. Of information I'm going into battle next year so Andy's taking it over next year and then there's a third course called the analytics, lab which is all about project. Based analytics. Projects. So we team. Up with a bunch of different companies. From around the world they give us very large data. Sets and, in. September, the students make teams and they work with these you know 20 million, items. Data sets and they spend the next three months, analyzing. Them and in December, we spend the whole day having. The students present back to the companies what, they discovered, and every. Single one last year in almost every year they significantly, outperformed. What the companies were able to do with that data, and came up with new insights this. Is important I taught at Harvard Business School for a decade before I came back to Sloan and I, looked back on that decade and all I was encouraging, students to do was, be incredibly fond of their own judgment after reading an eight-page case about a complex business situation, and go out there and I was populating, the world with hippos, well. But this is precisely the thing that's interesting is you would think the classic, MBA. Methodology. Popularized, by Harvard is the case study and, you think you would entirely, change, that methodology, based. On the advice that you're giving which strikes me as extremely sound. Working. On it. Actually. If you were to look at it wasn't it's not just our course if you looked at at the Sloan School and, my teacher we, are much, more project-based. Data-driven, and moving, away from eulogists. Or the traditional lecture, style or the traditional case style now much, more of the projects are we call them action learning where you guys are working with data and putting in place some of these principles in real world situations do. You have any and this is a little bit off the cuff but LinkedIn, is also, trying to help solve. This problem because you know please speaking whether, it's information for the newsfeed whether it's you. Know LinkedIn learning. You. Know kind of and Linda it's, it's how do we have people. Continually. Adapt, the, new skill sets to be successful, in any industry right it's kind of as a primary, mission across all of LinkedIn in. In, this intersection, between the stuff that we're doing in the stuff that, you guys are kind of highlighting, in. Addition to like obviously, saying look be, experimental, orient. Towards data, use. Judgment but, factor, data into your judgment right fundamentally, don't think it's my intuition, is the important thing as much, as my, judgment. Of data is the important thing is there, anything that you think that LinkedIn could be doing that. You would you would think of and this is a little bit of a kind of an on the spot question, and so it may be a, but. That's something that we also treat is a very fundamental mission, I don't. You guys get swelled heads but you guys are maybe best positioned of anybody to address, this kind of problem you have data on all, the jobs people are doing how they move from what together what the career ladders are what the skill gaps are I think you could be doing a lot more you can are you're getting a salary data everything but. You are very very, well-positioned to understand, these skill gaps which we haven't talked it much about it here in the panel but in our in our books we talk a lot about you, know the way that there's a growing mismatch, between the skills.
That Are needed the human capital versus, what the technology or what the people are doing in fact you, know quick anecdote we, way we first, met Reid was at Oxford University. Some. Years we wrote a short, little, bent pamphlet, type book called race against the Machine and we, had a debate at the Silicon, Valley comes to Oxford and it was very much around this topic about how there, was a single, single smartest, decision we made was to, let read me the anchorperson. It for our debate there. Were four people on team we were lucky the three of it we're all team, yeah. Exactly we were on this team and we want we won yeah we crushed them so that but, but but it's. Very. Clear to us, that, this, is a huge challenge for our society, and the, first part of that challenge is to. Understand, what exactly are, those, skill mismatches. Where is it that we are people, we don't have enough kinds of people and how you get that data well I'm, a roomful, of people who can work on that suit so I'd say you guys I'm glad you leave that chance you know Alan blue is playing. It wears out somewhere. Right here took off yes but but you know that, is that. Is a really big part of the solution to our society's, problem and you guys are well-positioned I think you could be doing more but I'm glad you're doing as much as you are and there's another aspect to this which is that, company. Like LinkedIn is in a really good position to help us solve an important. Really, fundamental. Puzzle. Or answer a really important question which is what, are the, characteristics of a lifelong, learner how does somebody actually manage to do that throughout. Their career, the dirty secret, about our. Industry, of educating, people is we have a huge, number of very deeply held opinions. And theories backed. Up by an incredibly, small amount of data and flipping. That around becomes, an important thing to do read, I was debating, our mutual friend Peter keel a while back and I. I. Disagree, with Peter Thiel on almost, everything. The. One thing he said. He's. Surprisingly easy to disagree with until you actually can he, demolishes, your argument, but, the one thing he said that I categorically. Agreed with was everybody. Says they're teaching critical thinking, most. People, are not teaching critical thinking so, how you actually go about building. That skill in somebody and having them be a having, them execute. A successful, multistage, career you don't know very much about that you guys do yep well, we're trying and we're trying to be data-driven so, let's switch a little bit to society. One. Of the things that's, that. I've watched with some consternation, over the last year is kind of this growing thing that's been called tech lash which. Is roughly. Speaking you. Know in you know in a kind of overly, simplistic diagnosis. You.
Have A growing, political. Storm. Around. Technology. Even, if a little bit less popular, most people like Google search must be like using, Facebook, to share things you. Know acceptor cetera but that growing, lash is a kind of combination of Republicans, who go at. Least the current cetera hub didn't seem to be trying to enshrine the past against the future in, my two, cents point of view and then. Democrat, it's going wait a minute maybe this was a problem for the election maybe there's a large large, companies, and large companies aren't normally our friends and we're listening to the media saying there's all these problems and so they're coming coming coming together and for. Me this. Strikes me as a, highly. Dangerous thing, for the future of society in the world because, while, I think there are challenges with. Technology technologies. How is a solution is how you improve it it's like how, do we get there not to say that everything has been perfect not to say there have them in the, errors, but errors and you learn from. What. What is your kind of your guys thoughts on tech lash what do you think society is you mewing what do you think politicians should do and what do you think tech company should be doing. What. Eric's, got a good three-part test I'm, gonna let I, mean say some so first up, are. The first digit is the tech is doing amazing things true sir sighs that's the important thing is that it's one, of the best things that that. Could change, so. Many things in health poverty, and so forth the old joke is that tech progress is the only free lunch economists, believe in that. Said, there's. Unquestionably, a tech lash and there's so it's not just you. Know made-up. Opinions, there's some real challenges. And I think tech. Companies have been remiss, and not getting in front of these challenges, and let me just I know Andy gave advice not to give a six-point, list how many of you a seven-part list really. Quickly though I'm just avoiding six. Very. Quickly so there's there's an economic challenge that so many people are being left behind median, income is. Stagnating. By most formal. Measures there's. A related thing that, big companies are getting winners, there's a lot more more winner-take-all markets, because of networks, Reid has written, a lot about this and, that leads to a concentration, of power a third. Thing is that there's, a cyber, risk, that is way more and more of our infrastructure, is digital. We become much more vulnerable to whether it's flash crashes, or hacking or attacks a fourth. Thing is that. There are a lot of developing, countries that have had, as a plan for the past 40 or 50 years to, do a lot of low wage labor, that, allows them to catch up and has work for some countries China Korea. And others but, that bridge, is being taken away, right now because, machines can do a lot of that kind of labor so there's our for economic, challenges they're also three other political. Challenges I'll touch very quickly one. Is is privacy, you know our phones the, internet of things we're just constantly broadcasting. A stream, all about us there's cameras, in more and more cities that have face recognition and, we're. In a new world we don't have the kind of privacy that we once did and that can be abused by companies, and by governments, there's. Issues around algorithmic, bias, as you have machines make more and more of these decisions, they typically, learn, from human decisions, and we. Were saying earlier humans, make, flawed decisions, so, that is a risk as well if, you embed them into the machines now anything I think that we can do better than we did before but if you just blindly have a black box that mimics what we were doing well, then you're gonna do what we were doing, and.
Then Last. But not least there's this whole issue around fake. News and amplifying, the echo chamber that, you know has gotten a lot of attention and it's not just fake news it's taking real news you know some bad. Thing that some antifa, or, whatever, person said and amplifying. It you know ten thousand fold or a million fold and trying. To get just you and him fight to get people angry at each other and that. And that kind of cyber bullying can be massively, amplified as well so there's at least uh at least seven, challenges. That I think are real and legitimate the, tech companies need to take the lead on in getting front of because if they don't it's, going to be the Republican Party in the Democratic Party the media and lots of other people who set the agenda and probably, not in ways that are a, necessarily. Friendly to tech companies but perhaps more fundamentally, not necessarily the right kinds of solutions for society yeah this is one of the very very few. Areas in politics, today whether is some flavor of bipartisan, consensus, right that the big tech companies are bad. And. Eric, gave seven, reasons why they might be bad but this is kind of big scary bad cloud, floating, around and I come across a lot of Hippo thinking, people just saying I'm vaguely. Uncomfortable. About, Amazon. Apple Facebook Google, Microsoft and, therefore, something must be done that is deeply deeply shoddy reasoning. And. I'm. Gonna I'm gonna do Eric's test for him he's got a nice way to think about this and says look let's if. Let's. Look the economic, impact of this in three ways how our competitor, is doing okay. You don't want to make the next social network you really, don't there's a lot of people, saying they're doing free R&D for the giant tech companies so there might be a chilling effect on competition what's, the state of innovation, these days one, of the things you really worry about with monopolist, as they tend to stop innovating, there's. A store in Seattle now where you put stuff in your bag and walk out the door and you're no longer shoplifting. I, really. Find it hard to say the state of innovation, is is dire these days those companies are among the world's largest vendors, on R&D. And then. The. Third leg is how are how are all of us - and how are consumers benefiting. Out there we're getting a cornucopia. Of, free, stuff and the biggest knock on Amazon is that the prices are too low it. Is really, hard to see how consumers, are being harmed by this so, if you're batting if you're batting 667. You know you should you should keep going to the plate, so. I. Want. To write back to one thing actually in the book because I just realized it was something that we, didn't cover that I think would be very useful. The. Economists. Advice across. This like part of what you're actually packaging, is there's, actually a bunch of good economics.
Work That, actually, in fact should be wrapped. Into business. Practice yeah what's the what's, the quick summary of that that will you know I, understand, the, difficulty the glass. Can. Economize for, a quick summary. How. That goes so each of the seven items, I. You. Know so, the short, answer is read the book yes, of course the slightly longer answer, is that, that in each of them whether it's the machine the platform the crowd there is some really fundamental economic. Work on how, you do decision, making how you do how you develop, platforms and leverage them how you leverage, the core in the crowd let me pick well I'll pick the platform, one and it may be you know and you can pick up the other platform that's interesting well we can you know don't do platforming mix and match I mean so, you. Know platforms. Is a can, be thought of a lot of different ways but one thing that we found very useful is this concept of two-sided, networks and work by John, to roll Nobel, Prize winners a bunch of other people are, friends Marshall Van Alstyne Jeff Parker have. Worked through the economics. Two-sided networks and some of the interesting things is I like one-sided, networks you know one-sided network is is a like, a phone. Or fax machine or whatsapp where the the more other people are using the same thing the, better off you are it's not valuable less other people are a two-sided never agrees when there's two different products, two, different groups of people and yet they benefit, each other so so uber, for, instance there's a an app that the drivers. Use and there's a different, related, app that the customers, use but and the more people that are using the uber app on one side the better it is for the people on the other side and you, can dear lots of other places where you had these two side networks but, it turns out that there are times when it really does pay to give. Stuff away for free to build the other network to subsidize you may even want to. Give. It for less than free so credit cards actually are two-sided network. It you know free a lot of the internet and originally. They used to charge the merchants they used to charge the customers, for, the credit, cards but now I mean how many people here get paid to use your credit card. Most. Of you probably you probably do you have loyalty points get loyal if you get free why whatever they are paying you if you're smart to use the credit card and now it's providing, a service to you but it turned out that the the elasticity. Of demand is, such that it's better to subsidize. The price of you to free or even less than free get, more of you using that particular credit card and then, they can make more money on the other side of the network charging, the merchants one to three percent fees and by, lowering the price on one side and raising it on the other side even below zero you can make more money than, if you thought of them as two separate, kinds of products and one of the counter into another kind of intuitive thing is that I think I feel like on or two is that, often. By merging, two-sided networks you, not only make a lot more money because you can play these kinds of interesting games it's, actually can be better for society that, there's a net, increase in consumer, surplus you're, creating more total value by growing the network so there's a lot more in the book that works do we even have some are publishers over, are publishers objections we put in some graphs, and charts some, lines they told, us not to do that but but you can see them graphically, somehow some of this plays out and that, gives some intuition because, it's not the answer that you just always give stuff away for free it's not the answer that you never give it away for free but if you understand the economics, of two-sided networks you can work through when you do it when you, we had a great discussion with with our publisher like we have to include a downward-sloping demand graph, in this book and they said there's no way you're going to include it downward-sloping demand graph, in this book so we we, fought that battle and we won that one another, read another example, of the Nobel prize-winning economics, that is playing out in real time right now there is mm-hmm.
Our. Editors eyes lit up when they saw that another. One is playing out right now and I think is a huge open question, is how, big a deal is this whole world of distributed. Ledgers and cryptocurrencies, and tokens, and icos and and. And, you're avoiding the word blockchain and smart, because. That's just one instantiation, of this broader phenomenon, and. It's I mean it is actually intellectually. Not only economically, but intellectually, the, most interesting thing happening in the business world these days I think how big it deals is distributed, alternative. Going. To be I. Came. Out not. There's. Something super interesting going on is this gonna be a revolutionary, thing that makes the company as we know it obsolete, I am firmly convinced, that it is not because, there's a set. Of economics, called the theory of the firm which is yielded just three Nobel Prizes so far at least yeah yeah Hart. Coasts. Williams, Holmstrom, Williams, system for for like a lot, of amazingly, solid, work on this that gets to the notion of what is it that a company is there for and how big it threat to that thing called the company is this, thing called a blockchain it's it's a super deep issue we. Wrote. About in the last bit of the book and I came away thinking this, distributed, world is going to have some impact it, ain't gonna make the world of the company as we know it go away actually, I agree with you on that and I'm, gonna do. Two more questions and, then there's a mic there if you want to line up for questions and I know there's an ability to do this online as well, and, so, and. By the way I'm kind, of plenty. More questions so people don't line up I've got a bunch of other things do but part. Of the the. Promise that we made Andy, and Eric was they'll, get a chance to interact with with, all that we really up, so first. One is in. The book the one that seemed least. Developed. As an area for recommendations. To companies was essentially the crowd was, essentially this new D. Centralize their I don't myself don't, really have them but, like the whole notion of oil you can you know you can deploy. A whole bunch of volunteers in Wikipedia you can use Linux as a way of of. Getting a whole bunch of volunteer developers, you, can have a. Non-controlled. System. Which you did a very good description of the Bitcoin blockchain kinda, like how does this work and what is the actual economics, of this work. But. The the the, map to if I'm in a traditional, industry other than know, this decentralization. Is coming and the people who figure it out are gonna have our huge advantage no we can do okay, what you tell the NIE sure you want to say that I think that's a particularly compelling, quantitative. Example, of how, people massively underestimate, the power of the crowd and they still do it and I think this this will help convince you there there are a couple point, things that I think any, decent sized enterprise could do to tap into the power of the crowd one is if you have a quantitative.
Tough, Problem, that you're working on, where there is an objective benchmark. Just, not like like hey is this a good idea or not you're gonna get junk back for that but if you have a tough problem and there's an objective performance benchmark, give, it to the crowd let, the crowd work on it our buddy cream Lakhani is a great scholar about the crowd effort, at Harvard Business School and we took we wrote up a study where, the National Institutes of Health said alright here's our baseline performance for sequencing, and the genomes of human white blood cells crowd. Can you do better and the crowd can went yeah. Alright give us a little bit on this they came back 14 days later and they took the run, time from ten thousand seconds to ten seconds 14 days, fourteen, thousand fold improvement thousandfold, improvement, and they took the accuracy up to about from 70 percent to about eighty percent the, total prize money offered. To the geek crowd for doing this was six thousand, dollars so, when we read this I went to cream and I said hey man it's the craziest thing you've ever seen he said it's about average said. If you can pose a geeky, challenge, and activate, a crowd to work on it you will get a quantum, improvement, the other let me give one more the other one we got was from a venture capitalist, colleague of yours Marc Andreessen, said look for all of history you had to commit a huge amount of resources upfront, before you launched a product out into the market now. Throw. It up on indiegogo is I mean just get a signal, about the demand, for this thing that you're planning to do that signal, was unavailable, to you before, and now it's pretty easy to come by emanuelle, yeah and even not just for little companies I mean in our book we talk about a GE, not. Knowing whether this ice maker was going to pay off and you know they were debating it and instead, of debating it they went ahead to IndieGoGo, and it was, yeah. And and. They found that they had tremendous, demand so but. The the lesson, takeaway isn't just that you get more eyeballs that's, part of it but it's more more fundamental, it's richer than that it's that the core the people in your organization they. Tend to be similar, to each other they have a certain expertise they're good at doing insurance, underwriting, or, financial. Analysis, or sequencing. You know genomes a certain particular way and so. They, could sort of you know there's diminishing returns doing more and more of the same things you throw out to the crowd you don't just get more eyeballs but you get people from totally, different fields. You know you get geologists. And petroleum. Engineers, and people are experts in fractals, and all these other things a fancy, term for have is marginality, you get people who are marginal, to the discipline, that pose the problem and many of those other ways of doing it just don't help at all but some of them you know for some of them it's just a fundamentally, different way of looking at the problem that is way, way better and you say you know what if we kind of reconfigure, if we really real able some of these variables, this is a trivial problem that we solved 15 years ago and you, know boom problem solved, and so, you tap into people who are just have more diversity of opinion, more diversity, of views and that's something that's much easier to do now and most companies way, way under well an opportunity, you can also imagine how thrilled the Corps is about, the crowd, this. Is not the oranges, department, typically does not love it when you show these results to, them so it's it's really managed, really difficult, and subtle, to start reorienting, the company and not making the R&D department feel they're gonna be fired next week but, gang we're gonna reach out to do this thing that we hired you to do we're, gonna reach out and there now there are platforms speak, of which to do this kinda quest you know you guys all probably know about Kaggle and the ability you can go there and and one of the things that calculus isn't just tap into this network of people but they help you reformulate. Reshape, your questions, so there's no form that can be addressed and you said upfront that you really need to have something quantitatively, and go after you can't just say hey how, could we be better you know that's not gonna work but, if you have a something, that you can define precisely enough, and people like Anthony Goldblum, help you or hell sail you know what that questions just not gonna work just go away but, often he will help you reshape the question of their team will in such a way you can tap into that power let the crowd interface, is a new capability that, companies might want to spend some time working on so I'm gonna punt because.
We Have a long line of questions I'm gonna punt on fake. News and advertising. Business models we might get back to it that was a minor interest that was my last clue but, I'm going to wait and do, these questions first please start hi. Mike. Mike, hi. Everyone Shankar Vanka common LinkedIn talent solutions, thanks. For, joining us professors, really appreciate it I was. At the C sales event in November both, of you spoke Allen was there loved it I Dec, see elements oh is. It okay go Brandon, not. Co-branded well enough I guess. One. Thing that came out for me work I fool these presentations. There's, pretty provocative and, you guys had I felt, pretty, compelling. Counterpoints, I don't think he was there to listen to that I think this audience. Would. Benefit from your. Take on that but Kylie, where he's coming from where China is going in terms of massive amount of data platform, e what he do in the u.s. is orders. Of magnitude smaller than what's going on in China what. Are the implications for the society, and the. Possibly. The reformation, of government globally. So. Like. You just pointed out our colleague, kai Fuli has. A lot of exposure to the AIE, efforts, in china and he, kind of frames as an arms race going on and he says china is extremely, well positioned to succeed, in this arms race i don't know how valuable it is to think about us versus, them situation. Here it doesn't feel to me quite like the nuclear arms, race but, let's take that framing and how, are we doing here I actually, had the chance to ask this exact question to. Condoleezza Rice and David Petraeus a while back when they were on a panel so I'm just gonna channel their answer, they said. Authoritarian. Societies, really. Are not great with crazy off-the-wall thinking. It's just kind of what they're not set up to. Foster. And real. Breakthroughs, in this discipline of AI are gonna require some, some fairly, radical since, innovative. Thinking, we also have a colleague back at MIT named rone s mo glue who co-wrote a wonderful, book called why nations fail and he, says look if you if you have authoritarian, States and and the and their economic, institutions, are inherently. Kind of extractive. You. Should not you should not be long on that so for those reasons while, I hear while kife who makes excellent, points, and they're gonna throw a lot of very very talented well-trained. People at this problem, I vote. On free societies, for dominance in these geeky iki fields. Just for fun I'm gonna disagree with and even though we, mostly agree, and I used to very much was, very convinced by drones work and, by that general argument I hope, I want it to be true but, I'm becoming more, and more worried about what the things that Kaku brings up and just to put some numbers behind some what you said, take. A guess right now the, amount of mobile. Payments on phones in the United States versus. China what do you think the ratio is and you know obviously, have more people we're. All more developed, we're richer so, you know if you took the number of payments. Through Apple pay etc versus, China what is it one two one five, what what do you guys think it is the only want to shout out a guess. Fifty. To one if you want fifty to one they have a thriving, mobile, payment ecosystem, that is just leapfrog knows I, was, just at triple-a I the the, AI conference. Last. Year the number of submissions of research, to triple-a, I from, American, versus. Chinese. Authors. Was, about even, this. Year, 50%. More submissions, to triple-a I from. Chinese than, American, research, triple-a I used to stand for the American Association for artificial. By. The way they changed the name of it is now the Association. For the Advancement of, our because. Because, you know they realized that this is not just an American thing so, and, there are many examples people, you know Andrew and others of pointing out that, in. A, lot of areas plausibly. AI. Researchers. In China are, doing as well or better than researchers. In the United States now disproportionately. You know to to. Agree. A bit with India and that disproportionately, that they were the, ones that triple-a I friends we're mostly an applied machine, learning sort of maybe not the more fundamental, breakthroughs. I'm not the right person, that. But. They were all tended, to be more applied aspects, of it and so there may be some element, of that, you do have, a harder time doing the more fundamental breakthroughs but I think this is a real. Challenge but I'll end on a point of agreeing with Andy that I hope, we don't think of this as a nuclear. Arms races and us versus them if they figure out a way to cure cancer or Parkinson's, or, have better self-driving, cars or whatever you know I think mostly.
That's Going to read down to the benefit of all of humanity likewise, as we make those kinds of breakthroughs and, if we can maintain an attitude that hey you. Know trade and research. And innovation, is something that's good for the world and not something that that's our bigger challenge, in this country right like maintaining, those adage and that's an economic and political and, cultural challenge, more than a technological, issue amen. Morning. Andy. And Eric thank you for joining us and Reed thank you for inviting them of, a two-part question two. Weeks ago there was a Bain & Company report. About the future of labor and labor trends, and they had their own three, around. Shifting. Demographics baby. Boomers retiring so it's gonna lower supply. Of labor to, is around automation, impacting, 20 to 25 percent of low and middle skilled workers which is gonna increase labor supply and three, is around the widening. Economic, inequalities. That those first you're gonna create so. First part is we'd love to get your thoughts on crystal, ball where you think, the. Workforce, is, going to be impacted if you have your own prediction, of what percent is going to be displaced, and. Just trying to get a sense for that too, is then I music. To hear to my ears when you talk about LinkedIn and your challenge, of how we can do more we've. Been thinking a lot about this is a company we acquired. Lynda.com. Got into online learning so. I want to get your sense of what can we do you talked about skills, gaps to, help these low and middle skilled workers who will likely be displaced by some of these technologies what can we do proactively. To help them I would love to get your thoughts in both of those thanks let me take the first one cuz that's actually the easier of the two questions, our. Colleague Bob Gordon has the best way to phrase what's going on with the labor force these days he said we do not have a job quantity. Problem, the robot apocalypse that, ends all that. Brings on massive technological unemployment. It's just not insight in the data at all we have a job quality, problem, and our, colleague David otter has done a beautiful job of of. Driving. Home the point that our job, creation engine used to kick out a lot of really good old-fashioned solid American industrial, age middle-class jobs and it was great it's, kind of down shifted, and now the sweet spot appears to be kicking out lower, middle class jobs more precarious less well-paid fewer benefits, all of that stuff what Eric and I spend a lot of time on is how. Do we tune that engine up how do we get the sand out of it whatever and give it the best possible chance, of kick of getting, back into that proper, higher gear, I personally. Think it's way too early to give up and say oh I need a universal, basic income because the robots are gonna take all the jobs that's a super overconfident. Prediction, history, should make us really, really uncomfortable about making those kinds of predictions about, tech unemployment. And they're a bunch of things we're not doing in the economy that that any economist, would say we should be doing to try to create the economic environment to, let the engines, of of entrepreneurship. And innovation do what they've been doing for 200 plus years which, is generate, a lot of demand for labor of all skills, that I for, me that's the homework what. Amy says 10x I agree with that we describe some of it more actually in our other book by some in this book about the kinds of policies. And. There's. A lot we could do to rescale, the workforce which is the second, part of your question I. Recently did a study with Tom Mitchell who's a professor at Carnegie Mellon University about. The, kinds of tasks that were most suitable for machine learning the kinds of tasks that weren't because we're very far from artificial, general intelligence so, that's you know not not doing that but there are some very narrow, stunningly. Powerful, things that machine learning. Can do really well but most he hasn't diffused through the workforce. Yet some of its begun to hit and we. Have like a rubric a set of tests you can apply to tasks. The nice thing is that the. Government. Has has, categorized. The US. Economy it's like 973. Occupations. And each of those has about 30 different tasks, like what does a bus driver do all the different steps so if you apply the rubric, to the tasks, which is a time-consuming. Process.
You, Can get a sense of how, the economy is changing and. Just. Summarize very briefly. Almost. No achey patients are entirely. Affected. By this wave of automation, but also almost no occupations, are not at all effective, and so our takeaway is that there's a lot of reorganization. Reengineering. We try a restructuring, as parts of tasks you know radiologists. You know reading the images the machines can do that better and better but, talking to the patient, coordinating, with the other physicians, you, know the overall plan doing, a history and physical all that part is still something that the humans do relatively better and even parts of the looking at the images they're different types of errors that humans and machines make, so, it's a much richer, situation. Where it's not all or none but, if you apply a rubric like that you can come up with a set of skills that are becoming relatively, more important versus less important, and that is I think a place where you know LinkedIn, and Linda and other tools can be used to, help shift, people in real. Time or maybe even in advance towards. Those kinds of skills. Andy. Eric, thanks, for coming and, Reid thank you so much for moderating so, well, so. For. Those of you don't know by the way we are collaborating with Eric and some, of his Co. Researchers on. Project, for the economic graph research program and we're gonna be hat we already have some results in working papers rolling, out but stay tuned it's very cool about measuring. How the returns to human capital are occurring to, companies. And to workers across the economy and there's more great stuff coming so stay tuned. Nadia, would be very mad at me if I didn't mention that but, but here's the one question I want to ask as you guys have talked about we, have all these amazing potentially. Transformative, technologies, are being rolled out here, and in lots of other places in Silicon Valley in, China else around the world and yet, when we look at nearly. All the economic, data they're. Not suggesting a world that's transforming, faster than ever it's actually, suggesting, us a world that's transforming. More slowly than it has for decades maybe, in a century, firms. Are forming more slowly, the, rate of occupational change has slowed dramatically, relative, to what it was. Productivity, growth is super slow it's, almost a uniform, message, which is kind of surprising what. Do you think is driving this disconnect between a very little economic change, and very, rapid, technological change and, do, you anticipate that gap is going to Eero how long do you think it's gonna narrow I'm curious for your thoughts, so.
We'll. Preview, we wrote an article that hopefully will come out as an op-ed soon up, called The Coming productivity, boom so the answer to your last part of questions we do think that it's it's in the pipeline it's because the, technologies, that we see is we walk around in terms of machines being able to do, image recognition as, well or better than humans or understand speech or just a whole panoply, of particular, skills, is awesome. But we also despite. All the investment, and say I'm self-driving, cars you don't walk around you don't see a whole lot on the street maybe a little bit more in this neighborhood than the, rest of the country but it's not something, that has fundamentally, changed the productivity. Of the country. Yet. Now, are people, like our friend Bob Gordon say well if you extrapolate from the past few years of somewhat disappointing. Productivity, growth into the future it's, bad news we, think that's a huge mistake you don't just extrapolate. Past. Productivity, actually has roughly zero correlation with, future productivity and we show that in some of our work the. Only way to try to make some predictions which is always a very risky thing to do is to, try to understand, what are the underlying fundamental. Technological. Drivers, and we will look at those we're, very impressed and the, the raises. The question well why aren't we seeing more productivity now and, we think the answer is that, this is just the nature of all, general-purpose. Technologies. And machine. Learning artificial intelligence, is a general-purpose, technology. That the technology that affects lots, of industries lots, of products, lots of services, earlier, GP, tees were the steam engine, at electricity, the. Internal combustion engine in each, case believe. It ought it was about twenty to thirty years between when, the technology was first developed and started diffusing, so when you start seeing a significant, productivity, boost now we hope we can compress that a lot where I think we're better at doing, that but we have to be realistic, that in order to take these amazing, technologies, and reinvent. Business models and come up with new products, and incorporate, them and and. Do it and do all of that we're, going to have to take sometimes gonna be a lot of entrepreneurial. Innovation now. One of the things that does trouble us that the guy mentioned is that. There. Are there's a lot more occupational, licensing, and other sort of sand in the gears of reorganizing. The economy, right now it's harder, to make some of those changes and many, governments, like you know that the city of Boston others you know they they intentionally try to read. Said earlier protect. The past from the future by banning. You know or putting, special taxes, on, ride-sharing. Services. And others so that, doesn't help this process of, reorganization. But, but as we've been saying several, times here we don't think that the issue is with the technology. So much as with the economic, and social institutions, and one of the reasons we're focused so much on trying to speed those up is to, is to overcome that those kind of dismal. Productivity. Numbers that we're seeing today and speed, up the time to address them let me just also just briefly it's been absolutely delight working with the team at LinkedIn. We were so fortunate that you guys have such amazing data and we, are able to this. Guy was saying figure, out how, the, value, of human capital not only affects individual, workers wages but in our project, we're looking at how it affects the, value of firms and a big chunk of the value of a firm actually, depends, on the human capital in that firm so therefore it makes sense for firms to invest in, their employees, human, capital, and not, just say well that's their. Problem or that's the government's problem and that's, part of the story that we're putting quantitative, data behind Eric.
And I are both really confident, that productivity, growth even with the measurement problems is it gonna improve sooner. Rather than later we've got an op-ed coming out to that effect I'm a lot less confident, that business dynamism, is going to turn around and start getting. Healthier, in America, people are moving less often there are fewer new businesses, formed entrepreneurship, is on the way down any measure, you want to look at at the dynamism, of the US economy is has, been on a long slow steady, decline, the scary part is we don't quite know why and one. Of our colleagues John halter Wagner has a great description he talks about it as death by a thousand, cuts which. Is kind of unpleasant to me because that means there's not just three magic things you do to turn it around but. I think part of our homework should be trying to figure out how to turn the corner and get get, our previous levels of dynamism back in the economy. Extrapolating. From what you all see around you all day is a deeply. Bad idea. Silicon. Valley is not America. I. Sound. Like a campaign ad for a candidate. That none of us would want to vote. Don't take the wrong message please, other. Guys. Thanks. For coming out first of all i'm. Thomas i'm a machine-learning. Sorry. I guess so, i deal with this quite a bit um, I have, two questions first part is, so. A woman, economists. Mariana. Met sacado she, talks about the cycle, of research and. How it relates to industry right so the, traditional model is you have the. US government funding, using public funds to create research into to do R&D DARPA. For example you, have companies private like, Apple Facebook Google, you know all the internet companies using, this research for free essentially and. Then building and making great profits off of this so. The first question is how do you think machine. Learning all that stuff is, being investigated, won by the US government for benefit, of all people and how do you think that's kind of shifting with the kind, of evidence, of these public conferences and more public papers and more information from private companies, the, second, question is in, terms of Education, democracy. Is all based on voters, being well informed and with technology growing at an exponential pace, how. Can the common voter be expected. To be informed, on these decisions and vote for policy, makers that are also informed you know we're having trouble with FCC and title 2 regulations right they don't understand what the internet is what. Are you guys you guys thoughts on how to better educate, people and, policymakers. Yeah. Let me do that let me do the latter because I think let, me try I think that's the harder question so I'll turn off take. The bullet on that one this is a really, really tough, problem. In. Both, the narrow sense and a broad sense how do you educate somebody, to be a decent consumer. Of facts is a really, tough problem and if anything again, do it to agree with Peter teal I think, our ability or even, our willingness to try to teach critical, thinking is, heading, in exactly the wrong direction a lot of what we're doing instead is teaching, indoctrination, which is exactly, the wrong thing to do, Finland, for example a parallel doesn't have much of a fake news problem, because, and they're right next door to Russia because, they have done a good job of educating their, populace. To be able to look at a claim and evaluate, and so well know that that that actually makes no sense at all so, part, so it's. Doable, but it's surprisingly, hard and I completely agree with you the urgency, of doing it is increasing. Because the pace of change in the pace of tech is increasing, and it. Is not the case that the, same side, of the argument always wins sometimes the crazy wins the argument the official position of the US government, is that climate change is man-made climate change human-made climate change is not, a thing we, the, crazy one the argument for a long time about vaccines, the crazy is winning the argument about genetically, modified organizations. And with all respect to Elon Musk, when he talks about summoning, the demon with AI man. I think that's benefiting the crazy side of the argument and I have seen what happens, when the crazy side wins it's bad news so, there's a lot of people in line so I'm gonna very quickly just amplify, what Andy said and then touch the other one and these are not. My men to be glib but short answers, I do think that. The. Answer is very much, around educating, the population so you know when we first started writing these books well how do we change things can we went to congressmen, and and said you know here are the things you need to do and they basically said.
You. Know I agree. With what you're saying but unless the voters want this you, know I can't I can't do it and I have to go with what the voters want so for better or worse we live in a democracy and, our, leaders. Will, to a large extent listen to us if if, the people. Are demanding crazy, things then they're gonna be pushing that way so unfortunately. It has to start with getting, people to you, know read our books and listen, to you and do it and do the kinds of things that we think will, change. The conversation in, that direction so that that's kind of hard a lot of hard work but, it's it's a I think that way that it has to work on exercise on the first point about, this cycle that Professor mosquito. Described. I think that you know she's got a very right, that that. Investment. In basic R&D has been a huge. Growth engine may be the most important growth engine for a lot of technologies. And it's, really I think economics, 101 that, there's. A public good for basic. R&D that, is not something that companies, will invest in enough on their own they don't have the private incentives, to do so so, we have to have government investment, in you know the internet and other basic artificial, intelligence, and then, and and fund it in universities. And education, and government labs and elsewhere, and. Then I, think it's great when companies, pick up that thing and develop it and create. Products, and services, and they have a private, to do that and they, become you know millionaires, and billionaires, I bless them that is that's the way more or less the system is supposed to work that there's an incentive to pick up that technology. And commercialize. It and then ultimately you know to close the cycle then you have to have a tax system that takes some, of that money that's been generated and uses it to pay, it forward for more R&D and research so that's that's the economic, 101. Paradigm. And I buy into it thank. You. Hi. Thank. You so much for coming, and talking to us so.
I Think we've seen that machine learning is it's easy to apply it to problems where the objective is, clearly standardized, the, inputs are standardized, there, are a lot of human problems where that's not necessarily, true I think, one really good example is, the, world of all mine dating right now there's been a lot of approaches, to introducing. Algorithms. Collecting. More information, from people on both sides to. Create matches. And. There has been some success but there's also been a lot of papers, that show that these. Algorithms are not significantly. Better than randomly, matching, people. And. The result is getting drunk in a bar and seeing what happens. So, the reason I think a lot about this is I'm. A p.m. on the jobs team and. The, match between an, a candidate, and an employer is actually. Not that different I think we like to think that there are a lot of standardized, criteria, skills. That people are looking for but. What we've found is that recruiters, a lot of time use judgment. That they cannot explain, -. To, talk about a culture fit or someone just clicking kind, of similar to an online dating sometimes
2018-02-27