[inaudible discussion] nod [inaudible] discussion [inaudible discussion] -- timely we're having this conversation today. a little trivia fact for you. today is the ten-year anniversary of the debut of iphone 1. so with that in mind, we are thrilled to be welcoming back to

the stage -- we are welcoming back to new america, andrew mcathe and erik brynjolfsson. they're back testimony with a followup. that book is "machine, platform, crowd," harnessing our digital futures which we have copies of for sale forking the conversation, shy mention. by further introduction, andy this principal research scientist and erik is the director othe initiative on digital economy at m.i.t., and both of them are the top headed scholars on how technology is changing business and economics. joining them on stage, alice son griswold, a reporter where she covers the economy, another sharing economy and strata conflicts. she comes from slate magazine where she wrote about business and economics for in the money box column.

so we have copies of erik and andy's book for sale. following the conversation and they will stick around and autograph them for you. without further adieu, andrew mcafee and erik brynjolfsson, and alice son. >> thank you for coming. tell us bat your book. >> we'll dive right. in you start. >> the book, called my been, platform, crowd ," available for sale, and people drive down the resale value of it by signing it for you arewards. this is about three great rebounds. mind and machine, product and platform, core and crowd. we couldn't fit all the words on the cover so just used the last three and that's the direction that the world is changing right now. not all the way, but partway from decisionmaking, moving more from human minds toward

machines, both data driven decisionmaking and artificial intelligence, moving from product towards platforms like the five most valuable companies on the planet, apple, amazon, alpha get, google, microsoft, and facebook, and moving from the core to the crowd, the core being the people at the center of the company or organization that traditional hey have made most of be important decisions towards the crowd is people out of it and now can include billions of people connect evidence with a digital network in a way they never were before. >> we should tell you the book came from. ways heard, erik and i wrote a book that came out in early 2014 called "the second machine age" all about the period of crazy tech progress we're living through, and as soon as we publishedded we notice this really interesting phenomenon, we would good to conferences and talk about the book and then we would have really interesting hallway conversations over and over again, and the same hallway conversation.

somebody would come up to and say, i run company x, and i believe the story you're telling, now what? what do i need do differently? how do i need to think about my business model, the industry i'm competing in, this world we're heading in. i know thursday are changing. need a guidebook for this, and so the more we heard this the more we real realized this would be the next book. the central problem ways didn't know the answer to that question when we first start hearing it over and over but we turn that bug into other feature and this he can intellectually rich territory to write a guide book to the new world. and in addition to doing what agency does which is read research and digest papers and do all of that, but we do that is a little different and that's much more fun is we go out and talk to the alpha geeks, this is a term we learned from our

friend, tim opretrial reilly, in -- o'reilly, the future of the world and we'll talk to them and our homework was to try to distill down what they were saying and come up with some kind of framework or way to think about the changes that are happening so quickly, and that's where erik -- the three part structure that erik just described came from. we kind of heard that machines were getting ridiculously more powerful. couldn't leave out the artificial intelligence and machine learning revolution. we kept on seeingwards companies appear and disrupted industry after industry and they all looked like platforms to us. we kept on seeing these examples of companies and individuals and things like -- look at dish wikipedia, very diverse people though crowd was a powerful phenomenon. so that leads to three-part structure in the book.

>> so, i think one of the an an neck e neck -- an net dote is about alpha go. er who here is familiar with the alpha go story? >> how much -- a lot of people. >> okay. >> so those who aren't, do you want to explain what happened? >> sure. alpha go is an example of a machine learning algorithm that wag trained to play the game of go. the ancient game of go is that sort of a little board game but vastly more complicated than, say, chess, which was beat by machines back in 1997. there -- it's-there are more possible moves in a game of go than atom inside the universe. and the tradition principal programming techniques of trying to sample even a tiny fraction of the moves is completely infees able, and for that reason, as recently as 2015,

computer scientists were surveyed when the thought machines would be able to beat humans at the game of go, and they predicted, probably around 2027. that was the median estimate. these were insiders, we polled this poll of artificial intelligence experts and asked the question. as some of you know, a machine did beat the world champion in go just last year, and they did it by using the new technique, not trying to exhaustively search the move buzz by looking at the board and seeing patterns in it. that's the way human does it. humans when they mail the game of go, they don't even know how they do it. if you ask them, you you play go, they can't explain it. they just say, look at the board and i see a move that seems right but i cap explain how. i can't describe the strategy, can't tell you what i'm doing. okay can't explain it, how do you program it, code it? but the new approach to machine learning where the machines look at lots and lots of examples of successes and failures, the

machines can figure out the patterns on their own, and by applying a technique called deep natural net and reinforcement learning the minnesota learn the game of good to the level where they can beat the humans at the game of go, and now the human champions, when they see what the machines are doing, they say, you know, we have not even touched the edge of the game of go. we have been studying this for 3,000 years but the machines so far ahead of us we have no conception of what is possible. >> the reason erik and i think we're still underestimating the impact and to power of machine learning, despite the fact we all read about it in press all the time is exactly what he gist said. here is the domain of intense human study for three millenia. we build a body of what way

thought was exhaustive technology, this machine shows us how much more headroom fliss this particular domain. that's amazing. when we think about all of the other incredibly complex things that we're trying to get good at, whether that is solving the mysteries of cancer, whether that is unlocking the secrets of protein folding, that is figuring out what to investy, whenever that is dealing with the silly complexity over the human genome we now have a colleague for that work that is not recommended be the knowledge we humans possess, might be able to show us some ways forward. >> this last points is really the key one and one reason we call it the second wave of the second machine age. the second member age is this idea that machines starts doing mental tasks the way that the first machine age that started doing physical tasksment but the way for -- the past few decades we taught machines to do fundamental tank is we programmed them. codified our own knowledge and encode it so the machines can

do -- how to do tax preparation or a spreadsheet. stuff we any what to deand unhad to be very precise and the machine would do the same thing. that's great. that generated trillions of dollars worth of value. that's great. but there's so many problems that it sound paradoxic cal that we don't know how to do them. i wouldn't know how to code how to recognize my mother's face or how to ride a bicycle. these are thing wes just know how to do intuitively but can explain them. we know more than we can tell. with the machine learning we just described, we now have opened up those possibilities. these machines are learning on their own how to solve the problems, not by having a human teach. the seven by step but instead of giving them examples what works and doesn't work. this is a cat, a dog, a human face, this is the word yes or no in french or in english, and the machines -- give them enough samples will figure out the

pattern on their own. it's really a remarkable transition that has really become important the past fewer years, since he wrote the last book, the second machine age. >> there was an example that went viral on twitter a while back. could the machine tell whether it's a blueberry muffin or a dog? because it had the little dots on the face. one thing love about that -- >> another ones that came out at the same time, the labradoodle the kentucky fried chicken. it's surprisingly hard to tell difference. i was like, dog, dog, chicken, dog, and not getting it right. >> a story that works better if you look at the picture. rite now you don't see that someone would accidentally bite into a piece of labradoodle. if you see the pictures, they can be hard to tell apart, but we humans can often do it and what is really remarkable is that seven years ago, machines had about a 30% error rate on a

big database called image net. that was using the best technique. now the error rote is lower than five percent, which is less than a error rate of humans on the same data set. so we have kind of crossed an infection point. jot just didn't -- inflection point, when you raise water past the boiling point it's a fundamental transformation. every machine can do a task better than a human innings you're an entrepreneur or manager and have to make a decision, who do i assign this task you pick the one that does the job better. >> one thing we cook talk about ised the idea that in the future or even now, people will be learning from machines in addition to teaching them, and one thing liked about the alpha go story is that after the world champion was defeated, he said, so beautiful, so beautiful, about the moves. it wasn't something heed ever conceptualized before and that's something we can sunny a lot of field its. >> seeing it over and over and

you're exactly right about people learning side-by-said misminnesota. lisa dahl, a guy who was bet my alpha go. a reason that where was a downward lip in the korean economy. it was big deal. the think two really important things happened along the lines of your question. one of was that sedal was beating 75 offers of his top level human opponents before he played alpha go. after he got done playing with alpha go he was beating 90% of his human opponents so he made a big quantum leap in his game, already very high level. the other thing that happened is the salutes go boards boards ann korea and asia skyrocketed and little kids started playing the games at much higher volume than before. so one thing we hear, isn't taylor wish no longer the best

at this particular activity and doesn't that reduce human value or dig it? i think that's dead flat wrong because if we define ourselves by the thing we're best at, that's to silly thing to doment we have this neat thing happening that people want to be part of. if technology is part of that neat thing, that's good. >> that do you think? >> i think it's good. let me -- understand scone pint that andy made. he got better because he played a bun of games with the machine? no it's because this machine came at the game from a fundamentally different way. when they saw some of the moves the machine made, the go experts said, i at first, my god, that is idiotic. >> >> there's because ear. >> every beginner knows not to do what the machine just did and then a few later the mon won and they are like, oh, i see what he did there that's really

innovative. so for 3,000 years humans have been learning from other humans, from their masters. if if the made the move the machine made their mast we are slap their wrist and sky, don't do that. but the machine came at it with such perspective they have a different way of thinking if you want to call it that and came up with innovative approaches and not just true in the game of go. it's true in all kinds of problems. the machines right-hand side just better in certain ways but most importantly they're gift in certain ways and they come at it with a different perspective and the possibility that humans and machines can work together and help each other because they're not just two clones of the same thing. their two different kind of intelligence. >> what is cool about the stories -- one thing that is cool is that you probably needed to be a player somewhere knee the level of seddal to get the insite why alpha go made the move it did. you need to be good at this domain to start to unpack why the machine was behaving in this

way. the machine was mute why it was doing what was doing. we need really high level humans to look at that and say, think i know what's going on here. let's test that idea. >> where are some companies parts of the economy that we're already seeing this. ... so people come to the website and have a chat with sales reps, is this course right for you, human sales reps.

stay tuned. some of them are better than others and answer questions effectively and convert people and they are a good course for the sake of it. others not so much. they recognize the transcript of all these chat become the data for supervised learning which is giving examples of failures to a machine and let the machine give the transcripts and their machine learning outlets are patterned, successful sales reps have ways of answering questions. at that point they do not turn it over to the boss because there are too many unstructured questions that there is an 80-20 rule, very common questions that account for a big percentage of queries and a lot of humans

interacting with incoming traffic but one one of these common questions came up, here's a phrase you might consider using. it has worked well in the past and humans pick up on that and the not so good sales rep, performance improved dramatically and the sales closing rate went up by 50%. a tremendous improvement but part of the with the ability to combine humans and machines in a way none of them would have been successful on their own. >> in some cases we will automate work, from mind to machine, that is the appropriate thing to do but naïve to think that is the only force going on. we love that story because it shows fans of technology and automation realizing humans involved in this process but we amplify their abilities or turbocharge their abilities with

smart technology in the background. >> each of those, we are not seeing turn the dial all the way to one extreme. we are saying you got to about the balance. it makes sense to automate the platform but in most cases you need the balance and the balance is more toward machine platforms than most companies or individuals realize but not as simple as the first part. >> this question comes up a lot, and the advent of machines, we were talking about, would you like to quote it? >> the only difference between

the unlinked series of observations is on the part of a songwriter. he is lowballing really badly because it would be trivially easy to program a computer, might rhyme and might have words associated with love and loss in him. we are incredibly confident that the lyrics would be not making a lot of sense or activate any emotion on the part of the rest of us. great human lyricists have a barely linked observation in a way that resonates with other human beings and that is the skill we have not seen computers come up with yet. there's a concept of linguistics i find incredibly helpful here called the intuition of the native speaker and what that means is if i look at any length or complexity in english, i can tell what is grammatically

correct or not. before my brain processed its meaning or know what it is trying to convey, that is kosher or not. i speak some french but i can't do that infringe to save my life, english is a second language, they can do as well as i can or any other native speaker can. we humans have native speakers intuition about the human condition, the social world we created, the physical world we move through. trying to teach computers to have that native intuition. if we do that i believe that is decades away. they are second-language speakers about the world we have created. a lot of confidence human ability to invoke an emotional response via words and language, if that is automated quickly, they are surprised. >> it is hard to tell apart from

the music of famous composers. >> not hard. >> the interesting question is what is the difference between music and lyrics and what about that makes it automatable and not? >> i am not a musicologist but the aesthetics of music are well known. music, we know what sounds good to our ears and what sounds like sirens clanking on the new york street. we want to encode that stuff and we know the rules of these kinds of things so relatively easy to spend those out. >> machines that are doing a good job are not encoded with rules or courts. there are patterns that emerge. we don't have to know anything

about this deep structure to see these patterns. machines are good, shadow pattern recognitions but the lyrics to be evocative, you need to understand the meaning, not just two words adhere to each other in sequence. if you want to go beyond that you need to understand the human condition and the machines although they can recognize patterns in words, they do not really understand the words. machines can translate one language to another, but they are not understanding what was going on. machines can be good at certain video games, quick reaction shooter video games, shoot the missile, great at that but if it involves deeper planning and structure they are hopeless and i think language is like a complicated videogame and music

more like pattern matching. >> we include in the book an excerpt from a machine learning system that gets trained on jane austen novels and generated jane austen pros blues not that it is gibberish that makes your head hurt. it is unpleasant to read. unless we can make a go 180 ° different i don't think that will happen anytime soon. >> it is often said when you look at jobs and what jobs are most likely to become automated, the divide isn't high school/low skill so much as high-frequency tests versus variable unpredictable tasks and -- >> the broader lesson is machines are not just good but superhuman at certain types of tasks. they have been superhuman at arithmetic for decades so we shouldn't be surprised and they are getting superhuman at pattern recognition and things

are repeatable and you get the data, able to train the system if they involve more complex planning, creativity, those are things that are much harder to have a machine with the current wave of artificial intelligence. we talk about what is available circa 2017, and all sorts of things breakthrough in 2018 or 2080. based on current technology, basically taking a bunch of input and massing them into outputs and figuring out the really good at that, not much complexity. >> that second point you brought up is becoming out of date because i never would have thought of medical diagnosis as any kind of routine, repeatable, low-level work you would hand off to a machine. it is an incredibly important

and difficult and subtle and complex pattern matching exercise. machines are better at those things than we are and they have come out of essentially nowhere and reached superhuman levels of performance. i medical diagnosis is in the sight of a highly automatable task that should apply more machines did not appear to be the case 5 or 10 years ago. >> part of crossed that threshold. if you look at image recognition, recognizing dogs or cats, you can also apply it to medical images. here's a picture of a lung. which has cancer and which doesn't? human pathologists and radiologists spent years of training to learn what to look for. you train a machine learning system to do that, rapidly learn not only all the markers taught in medical school but a number of new markers the no humans noticed before. >> what is the accuracy?

machines are rapidly equipping human experts in their own subjects. >> this is a number that is changing as we speak. there has been significant groups, an article in nature in february of this year. the example i gave from 30%, 5% from may, voice recognition, went from 8.5% error rate to 4.9% error rate but it is not just the improvement but that improvement was done in the last 9 or 10 years, the past 10 months since -- 11 months since july of 2016. as we speak they are improving it more and more and a lot comes from taking broadly the same set of algorithms but a lot more data. a lot more examples and a lot

more power. >> with medicine the studies we have come across our kind of conservative. we want to be circumspect about these things but they say are properly done machine learning system will be at human level performance, experienced diagnostician performance, that is lowballing it, they are better than the average doctor in a lot of cases today. >> where are people going to encounter this technology? is it alexa? >> you do get those errors. one example is voice recognition. that has improved a lot. it is not perfect yet. it still makes those kinds of errors that you just described but it has been measurably bladder to the point that i dictate rather than type a lot of the text that i write and a study out of stanford dictating as opposed to typing is three

times faster and more accurate for the applications they looked at. we are encountering some of them. the leading example that has been most obviously possible, to serve up next to your searches. if they can figure out what you are likely to click on versus what you are not likely to click on is worth billions and billions of dollars. that is something we encounter quite a bit. >> we are throwing this brainpower and computing power at ad placement. it is being translated into all kinds of other domains. the main thing before coming across this cutting-edge technology is devices like these, speaking into them. having them understand what we want, getting decent quality translation, type in a few letters and figure out what you

are looking for. you may have noticed, those who use facebook with 2 billion other people as of yesterday it will offer to label the names of your faces of your friends. it recognizes who they are, using this technology, that is something you couldn't do a couple years ago. >> what is the line? when facebook offers pictures, i find that creepy. >> first time somebody else tagged me in a photo and put that on facebook with years ago i found it deeply creepy. creepy is increasing boundary and i think creepy turns into familiar very quickly. the analogy i use when i was learning to scuba dive, first breath underwater was abjectly terrifying. you are not supposed to breathe underwater. the second was weird.

>> classic arthur c clarke in, any sufficiently advanced technology is interesting visible from magic. maybe that is a little creepy. companies are pushing them out need to think about how people are going to react. there are examples where they cross that line, they reveal that they know more than they wish they knew. they know it but are not letting you know and that is something we need to think about. what can these machine learning algorithms in for about that, social network, android or iphone, and machine learning algorithms, and can potentially

be used to infer things. >> on the other end of public awareness. and concerned that some companies are moving quickly, it will cause and accident. it will cause more death, regulators to clampdown the program. >> we have a view on this, self driving cars are not going to be perfect, not anytime soon. there have been deaths but the right standard isn't or at least shouldn't be perfection. as you know there are 30,000 highway deaths per year with human drivers. self driving cars take 90% better, that would be 3000 deaths. as a coldhearted economist, less improvement, self driving car manufacturers get 27,000 thank

you letters. these cultural changes we talked about, we have to think about what our expectations are, where we like to set that threshold. >> we have self driving cars -- another one, scuba diving moment,the first minutes of autonomous driving are terrifying. it is not abstract come you are hurdling on a highway going down 101, the car is doing 60 plus and nobody is driving the dank. this was initially not cool at all and then it turns into extraordinarily cool and then it turned into kind of boring because the car was going down

the highway going 55. it was breaking correctly, and amazingly boring experience. >> in pittsburgh book, mine was basically fine, the whole trip there were a couple abrupt accelerations. the new york times reported, his car when he let him drive it didn't start. he will get the best lead for his story. this car was fine, the experience was fine. i pressed the start button, it won't turn on. >> this is really important, this question about autonomy as an aspect of turning things over to machines is an important question. technology is increasing highway fatalities, distracted driving deaths are going up in america. bus driving going like this, this is the worst combination,

texting technology. my view is bring forward the day we can turn this over to the machine, they will do a demonstrably better job and of how that will open life to the elderly, the blind and disabled. why would we hold this technology back? >> one of my favorite is what happens if a self driving cars coming down the road and someone decides to walk into the road and just stand there? how does the car negotiate that? >> it will stop. >> does the pedestrian have a newborn power? the car will just stop? >> pedestrians believe they have a great deal of power in new york. we don't think the power balance is a huge issue. >> to give you an example of the progress, the early stage, they are testing before they are able to recognize pedestrians and one to 30 frames, once per second,

now it is one. 30 million, a sense of 1 million fold. one reason we wrote this book it is not necessarily the technology that is the most interesting thing, and if the car is going to be trained to stop without question, it is not going to play chicken with the human and some might take advantage of our start bullying the self driving cars and think they will go in front of it and make the car flinch. maybe some people will do that and we may have to think about how to update our norms and expectations just like with social media and email and fake news and other things that emerged that were not

necessarily expected by inventors of those technologies so there is a constant evolution of business processes and expectations to use these technologies. >> fake news is a great example. it is amplified by our minds and brings up the great quote, we were arguing who said it first, a lie will get halfway around the world while the truth is getting its boots on. we had a colleague at mit about to write a book about how viral fake news is, it is incredibly viral. what do we do about that? can you convince people to start generating fake news and trying to pass that on? you probably can't but can the great social media platforms tweak their algorithms and used machine learning to stamp down the bouncing around of those signals? they could easily do that. what our society be better served by that? i believe it would but if they do those kinds of experiments,

people of propagating fake news are unhappy they can take your nonsense elsewhere. >> one question you raise in your book is the supposedly neutrality of algorithms. no algorithm is unbiased because they are created by humans so how do you reconcile those things? >> let me nuance that. not so much the algorithm. i just used it that way. >> natural understanding of it. >> the issue is the type of data to train these systems and that data can be biased. they can be biased because the data -- let me make it more concrete. you have an algorithm deciding who gets loans and who doesn't.

and people who don't get loans, if humans had a bias, racial, ethnic, gender, in the system, may learn those in its own decisionmaking, a biased algorithm, it is a concern because, the algorithm is generated on its own, and therefore we don't know exactly what is driving it. this is important, we have to remember humans are biased too. the driving car example, i know -- we talked about this.

we don't think the right standard is perfection that whether the existing system, the potential for being approved over time in a way that is often hard to do with reducing or eliminating human biases. >> i want to be clear about my beliefs. we can beat the baseline with algorithms very easily. how are people doing on these decisions? we are biased as hell. we have all kinds of problems. our brains are amazing, their bugy, glitchy, huge amount of well-documented biases. when example in the book, may be my favorite one. there's a careful study that shows you are going up for parole in israel. the single best determined and of whether you got parole was to -- the blood sugar level of the judge at the time -- this is a

bad way to make those decisions, no rational basis for defending fat. judges were completely unaware they had this bias. judges are supposed to be neutral dispassionate people, they are doing a terrible job on the societally important decision that was not a fluke, a demonstration of pervasive things we humans with the best intentions, still get wrong. i'm not going to speak for eric like we discussed this enough. it will be very easy to have less biased higher-quality decisions in domain after domain with a little bit of data and smart algorithms. >> let's shift, let's go to the big economy, the future of work, outsourcing of jobs, platforms. you start. >> one of the important phenomena in the business world and our workforce, a couple

things going on. the percentage of americans who have good old-fashioned stable 40 hour week, health plan benefit jobs are still around, the percentage of working people who have a job that looks like that is going down at a steady clip. technology is accelerating the pace at which that situation changes because we have the rise of the on demand economy, very powerful, rapidly growing platforms that connect people who want to serve with people offering labor and/or assets to provide that service, we know about the uber -- when we look around the world and tried to understand this issue, there were many opportunities out there. entrepreneurs building platforms, different networks of people and assets and resources and we have not seen the end of the platform of the economy.

>> this is one place we learned a lot outside the united states, china has an explosion of the oh 2 oh economy. this stands for online to off-line, the idea you can have an online device like your phone and it connects you to a set of online services. like the uber of not just picking up a car or renting a place but if you need your dog walked or your car washed or any service you want to have a person help you with you can click a button and like a remote control to the world people take care of it and it connects you to a burgeoning gig economy or on demand economy and one of the most rapidly growing phenomena not just in china but something we are bringing back to the united states, people -- we

talked -- it is another example. potentially at least much more fluid kinds of labor marketplace and capital too. when you think of the use of homes and rooms, air b&b, uber does both. uber cars or lift can be a more efficient way of using the capital asset of an automobile. those who have cars, most of my sitting idle 95% of the time in a driveway or somewhere. imagine if you increase the utilization by sharing it, to save 50%. that would be a 10 times improvement in that asset. >> one of the weird outcomes of writing this book is i go home at night in cambridge and passed by my car in my driveway and i

have this big expensive hunk of metal sitting there idle, greater the 90% of the time they didn't use to incense me in the past and now it is a colossal waste. the opportunity we have is to take the infrastructure out there in the world, everything from spare bedrooms to cause to trucking capacity to warehouse capacity and make it more efficient by having it used more heavily and it -- we should all care about efficiency. efficiency is massively important if we want to tread more lightly on the planet or make better use of our resources, stop cooking it over the course of the 21st century. we will awake away from consumption, we better get more efficient with our use of resources. the phones we are carrying around our unparalleled access, for making use of other assets and researchers in the world. >> now we have introduced this.

should we continue? we have lots of questions. >> you asked some hard questions. i am the president of the center of personal life in the city. we look for a wildland those things shifted as you describe, there is a dystopian and utopian way to deal with these things. this is more toward the utopian side given by the other side. >> drive away. >> thank you. >> you may have seen this past sunday in the new york times, someone you personally know,

fascinating editorial. did you happen to see that? >> i did. >> this is a man who ran serious pieces of google and microsoft in china, the top of the chinese peer amid on these things. he said we are heading into a situation in which every economy in the world has to decide whether they will become incendiaries of the united states or chinese economy because in fact these machines are going to require massive welfare transfers. what he was saying is we the chinese will do this better than the us. my question is if we head into a world in which we do in fact have, new york times headline,

roboohlist, if we do go there, how will world leaders deal with such a rapid transformation which is the conclusion of the previous book but not what we are talking about today. >> great question, thanks for asking about that. we did talk about some big changes. on the one hand the bounty, we were talking about that so far. not just more millionaires and billionaires but gdp per person, lifting people out of poverty in china and elsewhere. that is good news but we also talked about the spread. for the past 20 plus years there has been a growing gap between the rich and poor, growing

levels of overall inequality and a big part of the reason we believe is because of the way the technology has been used. we are neither utopian nor dystopian. we think we have choices and none of this will happen automatically. if we sit back and wonder whether it will be good or bad that is the wrong attitude. these tools can be used in different ways. taking a more extreme position than is warranted by the evidence. we are not in a world without work. we are on the verge of a world without work. we are the world of rapidly transforming work. there's plenty of work to be done but it is different kinds than before the result in the data is many people are having falling wages, 50% are lower than they were 20 years ago and

people below that have done worse and there is pressure on those wages, not that there aren't any jobs but -- there are more people working than ever before but we need to do a better job listening to the incomes of people on the bottom half of the distribution. i don't see a roboapocalypse or whatever the word is anytime imminence but i do see the biggest challenge that our society has ever faced over the next we 10 years in using these technologies to create more prosperity. i am quite convinced that it is possible to create that shared prosperity if we update policies or institutions the way our organization is done. this is not something somebody will do to us, something we get to choose was one reason we wrote the last book is to

provide some guidance on how companies and policymakers and individuals can make that adjustment. i will be happy to give details of that but i want to calibrate a little bit. there is a huge challenge, the biggest challenge our society faces today is how to use technology to create shared prosperity but we shouldn't there is a roboapocalypse, and it is the best thing, no guarantee we will do it right. >> we should be clear where another challenge is. the big tech companies are large and impressive powerful companies in a lot of ways. how do we know this? a map tells us this. if we take the total sales of those companies and compare it to the size of the global economy they vanish.

it is a rounding error. they have to grow for decades before they had any meaningful share of the global economy. big powerful tech companies, not saying be blasé and do what they want to. they demand little rationality. >> i have been following the work for several years and haven't had a chance to read the book. i downloaded it in my kindle a couple days ago. two questions. there is a lot of concern about the implication -- i will be interested to hear how to

address these concerns emerging about purely autonomous and systems, the other question, your thoughts on this, the conflict between centralization and decentralization brought by technology, if you think about distributed autonomous mechanization, a new concept in the last couple years, how are you looking at how organizations, policymakers and people who are entrepreneurs who may want to overturn the status quo think of using these technologies. >> the definition of a great question is one that we devoted two chapters of the book to so that is a great question. we spent the final two chapters talking about the second question you brought up which is

what are the implications of very new technologies, radical decentralization. we have all heard of bit coin. block chain is the ledger that stops bit coins all around the world. we have had ledgers in banks for a long time. the crazy thing about the block chain is it is a ledger that is accurate, constantly updated, appears to be immutable and is not owned by anybody or controlled by anybody, not controlled by anybody. the central bankers take that thing over, it will be really hard impossible to do that. we have not seen this kind of thing before in the history of the world. a pseudonym for we don't know who wrote the paper outlining the possibility in 2008, this whole world is less than a decade old. you bring up this great

question. where is this taking us? when we went around talking about this, some of the geekiest people taking us headlong into this amazing territory where we build organizations that are completely distributed and completely autonomous and stick code in those block chains, those letters, contract in them, we invent new currency and find new ways of doing business together and the concept of a good old-fashioned corporation with stable hierarchies and managers, people believe that is very rapidly in the rearview mirror. you bring up the tao, the tao is pretty cool, intended to mimic the mystical asian force of nature. this thing we built to be the first distributed organization. it fell flat on its face.

it was launched because whatever happens, whatever we do with this thing is cool and you have to abide by that. when hackers said -- this was so poorly written i can treat it like an atm that continues to dispense money after i have a 0 balance. he siphoned up a third of the total currency that had been deposited, if there is no contract, no core, we got to play by the rules, that is a completely kosher thing to do except the community that built it said it is not what we want to do and they unilaterally revoked it and put a reset button on it. it got me thinking these aren't as distributed and autonomous as we thought and had good old-fashioned management like companies have management. my conclusion is we are not in the post organization post company world at all and my bumper sticker is you got to

trust somebody. >> when we work on that part of the book we rely on some research by nobel prize winners the talked about how, much as you would like to write a complete contract that covers every possibility and codify it in the block chain or a piece of paper the reality is most contract, all contracts are on some level incomplete and there is a whole set of literature and because they are incomplete that is one of the reasons we have organizations, companies and ownership of assets that say what happens when the contract isn't fully specified? we don't see that a technology is going to resolve a fact of nature that we can't anticipate all possibilities in advance. the fact this technology could be used to create a they facto atm. we try to draw on those kinds of

insights to battle test the amazing technologies being described. let me briefly touch on your second point. >> we are low on time. >> artificial general intelligence. you heard of artificial intelligence. most artificial intelligence is narrow, the system that is superhuman can't play checkers or chess or tic-tac-toe or talk to you in chinese and other systems have narrow vertical capabilities. one thing researchers are working on his artificial general intelligence that like humans if it can speak chinese it knows what restaurants to eat in and all the other things you expect a good chinese theater to know. that didn't exist yet and is not likely to exist anytime soon. no one knows for sure because the breakthroughs haven't happened yet. i think it is great that some

people are working on what will happen to society if and when we get artificial general intelligence. profound implications for the economy, security, the military, what we work on is what we can do with technologies that are available right now. i think there is an amazing challenge and opportunity happening to the economy. hundreds of millions, billions of jobs being transformed. that is just from plain old narrow artificial intelligence, amazing enough and that is our focus but another set of people, who knows how far in the future? >> do we have time for one more question? in the corner. >> a bias from process getting into normative issues, we tried to figure out what is right and

wrong. and defined by the majority of the model and minority, will it be plato's republic where following up the last two questions. >> i don't think we will give those decisions substantially different -- we have in making them up until now. we muddle through, companies with decision, rights to organization, government and laws that overlay things, public discourse about these things. i don't that will change substantially. >> what is before us -- >> if we do that we are idiots. machines, the tools that we create and the society we create should reflect our values. if we outsource the value definition of machines, shame on us and we should have to live

with the nonsense that ensues from the that. >> this is one of the most important lessons. pablo picasso once said computers are useless, and machines are great at doing that but more important, not only will values continue to be important. and those more powerful tools, we have more power to change the world than we ever had before. that means what we decide to do is more important than ever before, doesn't matter what you do, now we can do amazing things, we can cook planets,

blow, create vast prosperity. we have to think hard about our values, do we want shared prosperity? or leave them behind those choices? the values question is essential. and something we need to focus on. >> one more question. i have a question about neural networks. don't know how much you have spoken about. can you talk a little about them and how they relate to machine learning exactly? >> most of the impressive developments have a really elaborate huge neural network.

>> and the neurons connect, and 1 billion connections and hard to know what is going on in the black box but they are immensely powerful as i said earlier getting input to a set of output sliding to any arbitrary function and in the past few years they have gotten so big and so powerful that they can solve many kinds of problems they never touched before and one of the reasons we are so excited about the first word of the book, the machine driving a lot of this. and i might be misremembering, the book was on sale.

[inaudible conversations] >> every month booktv on c-span2 features an in-depth conversation with a nonfiction author about their writing career. join us on september 3rd. eric at latest book is if you can keep it. 's other books include amazing grace and the best-selling biography of dietrich von hoffer. maureen dowd will discuss her book bush world. and the year of voting

dangerously. michael lewis talks about his books including the latest, the undoing project. he has written the big short and the new new thing. join us for in-depth first sunday of the month at noon eastern on booktv on c-span2. >> you are watching booktv on c-span2, television for serious readers. one of the things we like to do is talk about books that are coming out later this year. joining us is lawrence o'donnell. his new book coming out in november is called playing with fire, the 1968 election and transformation of american politics. mister o'donnell, why do you have the word transformation in the subtitle? >> so many things were transformed utterly in 1968. for example, that is when what is