>> I was born and raised in the Deep South, right next to Georgia. [Audience laughs]. No, factually it’s true, oh wait a second this is, ah it just okay, machines hate me, I told you. [Laughter] Yes, they just ah, they just messed up, there’s not one, one slide is missing, but look it was really Deep South of the USSR [Laughter] In the Republic was the origin, right next to the Republic of Georgia, yeah. So, um, and a speaking ah speaking of my homeland, it’s just a funny story that my latest book ah, ‘Deep Thinking’ was about ah, yeah AI, in my own experience ah, fightin’ working ways was machines um and ah, the book before two years ago ah, was called, ‘Winter is coming’ it was not a synopsis of ah ‘Game of Thrones’ [Laughter] it was about Vladimir Putin and the enemies of the free world. And while I was on the book tour, everybody wanted to ask me about chess and IBM deep blue. Now, when I’m touring with ‘Deep Thinking’ everybody wants to ask me about Putin [Laughter]. But I’ll, I’ll try to stick to the topic, I’m sure there will be a couple of questions ah afterwards, so I’ll be very happy to answer them. So, I’m not a politician, I don’t duck questions [Laughter]. Um so, um, it might seem strange that ah, ah the game of chess, ancient game 1500 2000 year’s old, g*d knows. It’s you know, um, um, it’s a perfect analogy of artificial intelligence, because we’re talking about, when we talk about AI, we should remember that there’s the letter I, Intelligence. And what could be better than chess to, to demonstrate that. Um, surprisingly a lot of people believe that chess is kind of the odd game, played by nerds played in the dark corner of café. Ah, but to the contrary when just you, look at Hollywood, Hollywood always used it as the shorthand ah for smarts for their characters. And look you know, aliens played chess, X-men, wizards, I can even mention vampires, not on the picture. Humphrey Bogart, it’s an opening ah stage of, of Casa Blanca, one of my favorite movies and for, for the chess geeks, if someone plays chess here, I can tell you, because I studied this position and I looked inside, yeah [Laughter]. It’s a, it’s a real opening a French defense that was popular in the forties. Humphrey Bogart was a decent chess player, so um, and um, I can also mention that, Alfred Binet, one of the co-creators of ah IQ tests, at the end of the nineteenth century. He was fascinated with the chess players minds, and he studied for years, again looking for some, you know, shortcuts, to, to the secrets of human, human intelligence. Yeah, and ah, it’s not surprising that also ah, um, game of chess attracted, ah, um those who wanted to build intelligent machines. But, as usual the first one, is, as you can see, the ‘The Turk von Kempelens’ Turk, it was a hoax. Um, it’s ah, it was a big miracle at the end of the eighteenth century, it was touring Europe and America, it beat some decent players and also some ah very ah famous but weak players like, Franklin and Napoleon. But of course it was a hoax, it was not a real playing machine it was an ingenious system of panels and, and the sliding panels and mirrors and a strong player was hiding inside. The funny thing is that today, one or two hundred years later, almost two hundred and fifty years later, the problem is the opposite. In the tournaments, we have other kind of hoax when the chess players are trying to hide a, a device in their pockets.[Laughs] So now you have to look for, for , for, computer hiding in the human body. [Laughs]  Inside. Ah, and the, ah, ‘von Kemelen’ story is famous but his second one is story circulator is very little known, um story about a mechanical device; mechanical device in 1912 was introduced. It could play only with one piece but it actually could make, ah, make a mate with the, with the Rook. Ah, but still it was, you know, it wasn’t, you can say the prototype, the first computer. The most, you know, interesting thing is that, um, the founding fathers of the, the science, like Allen Purin and Glochen Ohm. They were those, um, they were um, um had great interest for the game of chess. And they believed, yes ah, they believed that, um, um the game of chess could be, could, also, could be um, an opener for this ultimate secrets of human intelligence and if one then, chess computer plays well against the world champion or beats the world champion, that would be the moment, ah of you know, of, of revelation. Um, fewer, fewer, remember that, Allen Furick actually wrote the first chess program, all the way back in 1952 and it was a great accomplishment, but the most important one that there was no computer. So, it was, it was just an algorithm that um he used to play this one game ah, and, ah he acted like a human CPU.  So, um, now, it’s a’ it’s, it’s important to remember that ah, the founding fathers thought that, the way  AI will manifest itself is basically following the same path as humans. So it would be kinda the replica the way we, we work. Ah, to the contrary, contrary to the expectations actually moved in the opposite direction with, ah, with brute force. So, um, now, um, I entered ah the competition against machines in, ah, 1985. Um, you could look at this, this picture it’s the, it’s not ten actually thirty two boards and ah, I played humans, but as a matter of fact the real game was against computers. They have four leading manufacturers of chess computers, at that time there were some dedicated chess machines. Maybe some of you have them still, like you know, a piece of antique. Um and they had eight, ah models each and I played thirty two. And um, I won all thirty two games. Ah, but what’s very important, it was not a surprise, everybody thought it was a very, very natural result and every time I look at this picture and look at these games argh, I’m sighing, that was a golden age [Laughter] of chess. Machines were weak and my hair was strong [Laughter] [Audience applauses] Um, 1985, June, twelve years later, I faced just one computer, just one computer.  Ah, by the way, people tend to forget that match in 1997 was a rematch, because I won the first one in 1996, in Philadelphia. Um and ah, okay I won this match, but just to be fair, the watershed moment for the computer chess was not in 1997 but I would say 1996 in Philadelphia. Though I won the match, but I lost game one, then I fought back and I won three more games, winning the match four, two, two. But fact, the fact that a machine was able to beat a world chess champion in, in a normal chess game, that was already just like a big signing on the wall. The rest was a matter of technique, though I didn’t expect, ah IBM just to do so much work and just to come back, you know later with a stronger machine. But the biggest mistake, except not ask people stock options, [Laughter] you know two weeks of, two weeks of the evet and the ah, it’s the, it, it grows ah, from the 11.4 billion dollars in value. [Harsh breath] okay, but the biggest mistake was not reading the fine print. Because one of the problems in 1996 that I, I faced, while playing Deep blue, was, it was a black box, I didn’t know anything about the opponent and while preparing  for the game, whether it’s a chess game, or soccer game or whatever. You always look at the games and, and some strategies used by your opponent. Now Deep blue, no information, now, I tried to be smart and I said, for the next match, we have to make sure that, um, I will have access to the games played by Deep blue. They said absolutely, but the fine print said, ‘played in official competitions’. And of course Deep blue has not played a single game outside of the lab [Laughter]. So in 1997, I faced again the, the black box. That’s, [Inaudible] unfortunately [Inaudible] I won the first game and I, I lost the match. So, um, by the way, where were you hackers twenty years ago when I needed you [Laughter]. So, I think that looking at, ah front row, some of you may, might not yet been born [Laughter]. Um, so, um, um, the problem [Inaudible] states for me in that match was that, I still treated the match as the great scientific and social experiment. Because I thought it would be great, you know, just to actually check, at what point human intuition could be matched or even just overshadowed by the brute force of calculation. And again, Deep blue, even with this phenomenal speed, 200 million positions per second, pretty good speed for 1997. Ah, was anything but intelligent. Um, the way the Blue played, ah, that’s offered us no input in the mysteries of ah, of, of human intelligence. Ah, it was as intelligent as your alarm clock, though losing to 10 million dollar alarm clock didn’t make me feel any better. [Laughter]  Ah, um and ah, um, I just realized [Stutter] remember in the opening ceremony of the match when ah, actually the press conference when the, the, the um, the man who ah lead the project said, ‘its standard about the experiment, now it’s about winning.’  Okay, that, that was definitely about winning or losing so and I lost to the, Blue, of course I wanted to play another match ah and I then retired the computer. Um, okay, they, I said they killed the only partial witness [Laughter] Um, and eh, um, I was actually trying to find out, what happened to the Blue, just I couldn’t. But lately, actually I discovered, now it has a new career. It’s making sushi at JetBlue Terminal in JFK [Laughter] [Audience applauses]. I, I, I love sushi but I don’t eat there, yeah guys [Laughter] some I don’t know why, yeah, yeah. So, um, again that’s the, that’s the story was over for chess and that’s very quickly because, can as I’m sure some of you playing chess is, well it’s chess. We talk about Gor, about other games. Humans are vulnerable because we don’t have steady hand, we make mistakes. So even the great game played by the world; world leading players at the very top, ah a world championship match. Say fifty moves, forty five good moves, four great moves, one tiny inaccuracy, is inevitable. We, in the human game it doesn’t matter, in the fightin machine, you will be punished. Not losing, maybe, maybe not losing the game but definitely not winning so machine will escape. So, ah I just realized at one point that, it just will be a matter of time. Because we cannot reach the same level of vigilance and precision that is required, to beat the machine, because the machine does the steady hand. Again we saw the same at Gor, many, ah, ah years later. [Inaudible comment] Lately, machine ah, conquer the game of Gor as well. Ah, but again, it was just about the game of chess. And ah it’s a game proved to be vulnerable to the brute force. Um, but it’s, you know, it was not ah, it was not ah, um, yet um, an AI as had been reported by IBM and by. Few people remember the match, saying oh, it was the dawn of AI actually not. Um, and later on um, I played few more matches with the machines. Because, um when, when these days are analyzed these games, using the modern chess engines, ah it was quite a painful experience, probably back to the past, revisiting it and recognizing how poor, I played in this match. I can blame myself, but also the people were not strong enough, this is something that you may not believe. But a free chess app on your mobile device, today, is stronger than Deep blue. Yes, frosty, yeah and of course if you have a, a chess engine like, you know, Stockfish or Commodore and you have it on your laptop, it’s much, much stronger. And ah, I just, you know, run these games and it’s one of the moments, I think game five, just looking at the end game and the Deep blue saved game by miracle and everybody talked about the great escape and phenomenal quality of chess. Today, you put it in the computer, it’ll, it’s, it laughs. It shows within thirty seconds to a minute, depending on the strength of your, of, of your ah the speed of your laptop is that first it was a draw, people made a mistake then I made a mistake missed the Queen and then people save the game. So, it’s okay, that’s the, that’s, that’s the Moore’s Law, I guess and there’s nothing wrong about it. Um, and ah, ah those two more matches I played in 2003, they both match ended in a draw. Ah, I played, actually they forced me to wear the glasses, to play on X3D, as if playing a machine was not, you know, tough enough [Laughter] Yeah, ah, I ah did well, so I was quite pleased with, with ah, with ah, my accomplishment. But again, the story was old, so I knew it and um I just was thinking in the future. And just um just what gave me a good thought, just look at this picture. This is the kids, so you have the nineties; you have the, the beginning of the century, the century and then more than raise. So kids, you know they just um they have to look at it’s like piece of antique, my kids will not recognize it. So then there’s this more sophisticated ah, um, keyboards and know they just [makes sliding noise] they’re their fingers. So, um, what is important that is, it’s, it more intelligent machines make our task easy again. I’m, I’m telling you that so, you know better than anyone else. So, I’m leashing our human creativity by clearing the way of competitive and technical tasks. So, then I had a thought, um, how about combining the strength of machine and humans. And let’s use chess as the um, as um, oh, as an example, because in chess we have the resolve.  You know exactly where machine is strong, and you know what machine can do as well as humans. So I came up with, with a concept, um, that I called, Advanced chess, okay. Following a famous Russian saying, ‘you can’t beat them, join them’ [Laughter] Um, so I go with, Advanced chess, men plus machine facing another human plus machine. So, um, and in 1998 I played another elite player, Veselin Topolov from Bulgaria you can see his picture. Ah, we both had ah, again, pieces of antique. Ah, now the interesting thing is that, we did not do well, because we were not, we were not able to maximize the effect working with the computer. And I just couldn’t understand why, we’re great players, so what’s wrong with that? So, we didn’t do well, and the and the answer came later with the introduction of the so called, Freestyle chess tournament, on the internet. What I call, invitation for cheating. You can play on internet, ah, being connected to the super computer, you can have your own computer, you can have many computers, I mean do whatever you want. Now, as predicted, human plus machine always dominated super computer. Again, the reason is very simple, because machine compensates for our weaknesses, so we get, we get into a good position and you can switch it to the computer. So no more vulnerabilities of, of, ah of humans that can be exploited by the other machine. But the trick was not that is not the sensational result. The sensational result was that the winners of the competition, the first one, and it was repeated later, were not top players, but actually relatively weak players. Ah, working with ordinary machines, but having superior processes. And that me do, to make this ah formulation, which I think’s quite important because it’s, it’s hard to understand, it sounds like a paradox. That, a weak player plus an, an ordinary machine plus a superior process will be dominant in the game against a strong player, even strong computer an inferior process. Interface decides everything and ah, it’s ah its quite amazing that it’s just ah you, you don’t need a strong player, you don’t need Garry Kasparov. Just to be at the, at, at the side of the machine, finding the best move. And the answer is simple, because when you look at the relative strength of humans and machines today and I will go beyond chess, but let’s start with chess because with chess we have numbers. If you are aware, with the, with the ratings and the rankings in chess well, let’s ah, give you an idea. Um, when my, my top rating was 2851, when I retired, was, I dropped I was 2812. Magnus Carlsen was traversing 2800 territory as well. There are about fifty players or plus in 2700, early 2800 category. That’s, that’s, that’s elite, of the world of chess. Now today’s strength of the computer, it’s about 3200, now on dedicated software it will be 3300 to 3400. Now we understand why we don’t need a strong player, because a strong player like myself will be tempted to, push the machine in this direction or that direction. I will be challenging machines, ah, evaluations while to the contrary, I have to be an operator. So a decent player, that doesn’t have the same pride, the same honor as the world champion or strong player will be far more effective; in creating the human machine, human machine combination. I think this is this is [stutter] a very important discovery in chess. And I believe it goes beyond chess, for instance in medicine, we know today that, ah, ah in many cases, machines are far more accurate in giving diagnosis than the best doctors. So, would you, would you like a good doctor to work with the machine or a good nurse? That with full instruction and a little guidance but not will interfere, because if, I don’t know the exact numbers, but say the doctor will be, do 66% of cases and machine maybe 5%. Numbers are on either side, but psychologically if you’re a good doctor, you cannot accept it. So, one can look at the progress of computers these days is just basically we should realize that, machines [Inaudible], medical diagnosis, you name it, could be good at climbing at 80, 85 maybe 90%. But know that’s, that’s, that’s where we belong to, humans, the last decimal places. And it could make a hell of a difference, it’s like when you know we shouldn’t move it, just you know one degree difference in angle, and it could you know be hundred meters, ah, ah gap you know, wide on the target. So the same is here, it’s, it’s about our ability to actually channel this massive computing  power and just to find the right, right direction for that. So and um, so I still believe that with all the fears that machines are just going to replace us and just you know, it will be the end of the world and Armageddon. I believe there’s room, there’s plenty of room because as I said, it’s about human creativity and this unique tools and intelligent machines will enhance our creativity, unleash our creativity if we know how to use it. Um, so ah one of the, actually looking for the answers, sometimes you go off, off side, not in the ah not searching in the hall of science. Ah, but in the hall of art, and ah I found quite a good paradox that was ah, um, um, allegedly said by, by a great artist, ‘Computers are useless they can only give us answers.’ I think that its, it’s a piece of wisdom, then again you don’t expect you know Picasso to be on the side of philosophy. But I find it, I find it quite ah, quite ah, um encouraging. Because machines find answers and answers and an end and Picasso could not accept ends, he was an artist. It’s ah, he, he had to constantly re-invent; he had to re-invent constantly his art. That’s what we do, so this is exactly, where we, where we have to start, asking questions. Um, can a machine ask questions?  Ah, once I ah, paid a visit to the Bridge Waters, Byhalia Hedge farm. The reason, I wanted to talk to David Ferrucci, the father ah, of Watson. And we talked about machine’s asking questions and ah [Inaudible] he said yes machines can ask questions, but they don’t know what questions are relevant. Thank you, that’s exactly the point, so we are, we are still in the game. We’re still in the game, we still have a chance to move on and ah and that gives me a lot of ah, a lot of confidence that ah, the game, the game is not over. And, ah Um, just a few pictures so I um, um some photos from the future of autonomous machines and machines that, you know essentially program themselves. So the one picture there is Demis Hassabis and his AlphaGo.  Actually this is a problem of first machine that ah, that could be called ah a prototype of AI. As I said Deep blue, brute force, Watson still it’s, it’s maybe it’s a transition but it’s not AI. Now AlphGo is it’s, it’s a deep learning program that keeps, um re-inventing itself by looking for the patterns. While playing millions and millions of games. Now, I can tell you it’s the first time that we had link with, with real black box. Because with Deep blue for instance, if you had um, hundred years to spare and ah, and you be willing to look for thousands of miles of ah of logs, you’ll definitely go back to the original idea, why the decision was made. Now with AlphaGo, I don’t believe that even Demis Hassabis can tell you why version six plays better than version nine or other way around. So it’s a, it’s a greater accomplishment on one side, but on other side it ah it might be challenging because if there’s ah, if there’s ah back, so how we going to find it out. But again, let’s move, that, that, that’s a move in, in, in this AI is, AI direction. Um, and while I just you know I was, I, I was um, spoke at um, ah Googles HQ at um, Mountain View um, and ah they gave me tool of ah Google X. This isn’t [Inaudible] observation because obviously there are many challenges for self-driving cars and for other projects for, for the drones, flying drones dropping goods. But the biggest problem actually comes from not from maybe I’m wrong another problem, as big as a technical one comes from, regulations. And this is an interesting question, it was oh, machines are you just know you know killing jobs you know, that are replacing ah, humans. So what are you going to do? That’s called history of civilisation, that, that, that has been happening over a millions, hundreds of years. I think to the contrary the problem is not that machines are replacing, replacing human jobs, now on, on the intelligent, intellectual side. I say now machines are going after people with college degrees and twitter accounts. [Laughter] Not too fast, I think too slow, and let me tell you why because it’s a normal cycle we just don’t recognize that ah disruption means that the new technology, breakthrough technology, before it creates jobs it kills jobs. It, it, it, it renders whole industries redundant, obsolete. And then it creates new jobs, this is a process, this is a cycle. Now if you try to protect the agro, by sticking who is the old technologies by whatever, printing money or just creating some artificial advantages for the old industries, you made this process slower and more painful. It’s going to happen anyway, but the problem is that with so many regulations, we’re just facing that, many things are just, just been intentionally slowed down. Ah and um I believe this is, this is, it’s even a bigger problem than, than the challenges we’re facing. And, it’s psychologically, people say, oh how can we sit in the driverless car?  Really, I just looked in the back and just found a hundred years ago one of the most powerful unions in New York City was the Union of elevator operators.  [Laughter] Really, 17 000 strong You know it’s, it’s , cause people by the way technology, to push the button,  was there already, but people didn’t trust it. No, how can you get in the elevator and just to push the button? [Laughter] Aaaah. [Laughter] You know what, you know, why, why this Union died and why people switched on? Because one day they decided to go on strike, [Laughter] you know, strike? [Audience claps] and when people had to climb to the Empire State building, they decided, maybe you, you ought to push the button. [Laughter] And I’m thinking now just, you know, twenty, thirty years from now, our kids, our grandchildren, they say, how these crazy guys they were driving cars. Look at the statistics; you know it’s one of the greatest causes of human death [Laughter] how could they afford to do that? [Laughter] Um, and of course you know with this, it’s ah, it's pure psychology. So many accidents, we know, people being killed in car accidents, but if you have one accident in a driverless car, that’s a big story. Any, any glitch any mistake, you know made with AI, with new technology that’s a story, you know, front page of newspaper. But again, statistically, come on, that’s you know; simply don’t just look at numbers. Yes, I understand it’s bad if you are in this, you know in this tiny percentage, but as the, as the humanity we’ll all win if we just move forward, you know just without, ah being paralyzed by, by this, by this  field. Um, so um, and um, it’s ah, it’s a picture of the, of our security center and ah, um, you know, um, another portion is now, because we talk about fake news and we talk about cyber security. It’s, it’s, it’s a big political issue and there, there are many calls, so how, how’re we going to fight hate speech for instance? So, I do regular blogs for a boss, my new one that will be released in a couple of days, it’s, it’s about hate speech. As I say it’s the fighting hate, saving speech. It’s just; we should realize that, this problem did exist before. It’s not that they been invented, they’re been magnified, because internet just involves millions and actually billions of people in. Again, I think it's good news and we should just simply realize that it’s trying to stop it, trying to outlaw it, you know, it’s not going to work. Because you still have Putin’s of this world and you still have other, you know bad guys sitting elsewhere. That they will use our technology, created in the free world, against us. So, I think we should just embrace it, that’s my view, so I always say, it’s about us. The answers inside us, it’s about our own strengths, ah, and our own confidence and I say that intelligent machines will not make us obsolete, our complacency might. So, um ah I, I think that is just you know that we, we should just realize that again, there is certain limitations in this corporations of humans and machines, but there is plenty of room, there’s plenty of room, as it happened before, it opens new opportunities. It destroys the old world and creates a new one and sooner we move forward, the better we are. Now let’s just, now let’s move to more, just ah so, ah science fiction world. It’s an interesting paradox that when you go back fifty, sixty years, the science fiction was all positive, it was all utopian.  And then gradually it moved from utopian to dystopian youth. Ah, yes, we don’t want to hear about this future, by the way, it didn’t just happen overnight, it was a time when people decided maybe it’s too ah, it’s too risky. To ah, to ah do space exploration, actually it is too risky. You know, just imagine in 1969, 1969 when Americans landed on the moon, the entire computing power of NASA was less, than any computing power of any device in your pocket here. So, this device is a thousand times more powerful than Cray’s supercomputer forty years ago. So just imagine for a moment how much power we carry with us and how we use this. I’m not sure that, this, that Apple, um iPhone 7 is the same as Apollo 7, this has the same effect. And I think there are many great things that can happen if we start looking for, you know, just for the sky, for, for the stars again, Deep Ocean’s, there’s so many great things we can do. And again, we should realize that machines, they are offering us an opportunity to take larger risk. And um, I just wanted to end up on, on a positive note. Is it positive? [Laughter] Actually it is, now by the way, the, the picture in the bottom, you know it’s not ah Photoshop; it’s a real one, yes I was in the office of the Terminator in 2003, yeah. Oh, he loved the game of chess; his kids you know will have these mandatory lessons. Yeah, we played a game of chess ah; yes it ended in a draw very quickly [Laughter] And he was I’m sure he was so excited that six months later he ran for the Governor of California. [Laughter] And won [Laughter] Um, now you think why the picture is there, why it’s, why it’s, um why I call it positive, because you know, set aside the first movie, in the rest ah of the, of the series. Ah, it’s, it’s still just; you know,  Arnold who always is, always on the winning you know old, but not obsolete, beating ah newer machines. But actually it’s a combination of what I described few minutes earlier; it’s a, ah, human plus an old machine plus a superior interface. Dominating newest machines [Laughter] so I guess I, gives us a little of you know just self-confidence that working with machines and having the best interface, I’m sure you know you guys are just the best in the world who can do that, so this is how we move forward. And, ah and then for those who say yes, but machines will eventually get everything done, so this is, no matter what else, they will calculate everything, because  machines knows the all’s. They will calculate you know it’s not about calculating everything; by the way the game of chess for instance is technically called mathematically infinite, ten to the power of forty five, number of legal moves. That’s more than enough for any computer in the universe. Um, but the most important thing is this, again, is in the games it’s all, you know, it’s, it’s, it’s can be not calculated but machine can be always be ahead of humans, it’s all about playing by the rules and you know the rules are fixed. You know that machine, can you know just ah find the best, best path in, in, in, in this jungle. But now if we move into the, if we move into the just normal situation, now wives are you sure machines can be helpful all the time? Let’s look at a very ordinary situation, mundane. You have your computer running your budget and you are in the store, you buy a gift, an expensive gift and machine beeps, nah, you’re at your limit, machine knows the outs. But one more, slight change, you have your kid next to you and it’s his birthday or her birthday. Now how does it change the equation? It changes everything, it could be a wedding gift or whatever, I can start adding this little things that will change everything and I don’t think you can simply, you know incorporate it, into, into this equation. So, definitely we have you know, we have, we have a little room. It’s like asking the question, it’s and because the situation changes. And this is, this is something that you may call ordinary but I had something for in the movies, I have something more dramatic. Let’s look for something that is, ah, extraordinary. [Laughter] Empire strikes back, ah you remember this, this little episode? Hans Solo is, is, is, is, ah directing his ship into the field of asteroids and cp3O ah, cp30 is just aahhh, panic, the chances of surviving in this field are 3720 to 1, never tell me the odds. [Laughter] Now, this is interesting, it just, ah just, you know from human’s part to the other one, who was right?  Technically, cp30 was right, the chances of surviving were slim to none and maybe technically, being caught by Imperial guard was a better option, was it?  Because humans could recognize that even if technically for the computer eyes, the chances of being caught by, by Imperial guard, it’s all for the better, better odds, that wasn’t an option at all. So this is very important, that in many cases again both simple ordinary and extraordinary ah, highly unusual, so we still have room, we still have room to move on and just to, to make all the difference. Um, I’m saying that human leadership is still required, and sometimes, sometimes, it, that will mean, going against ah the computer recommendations. So, the essence of human leadership is not a question of knowing the odds, but a question of knowing what really matters.  Not just for the tomorrow, but for the distant future. Call it, human guidance, or you may even call it, human interference. Interference with our intelligent machines and I believe that will set the course for this century. Now it, sometimes it surprises people, that ah, I’m such an optimist ah, after about ah, intelligent machines ah, considering my personal experience. But, I, I’m an optimist, I’m a pure optimist by nature, I have to say, and I believe that you all too are optimist about the future of humans and intelligent machines. Because we should remember our technology is agnostic, it’s neither good nor bad and it could be used for good or evil. The machines will keep getting smarter and more capable and it’s up to we humans, to do what only humans can do, dream. And dream big, so we can get the most out these amazing new tools. Thank you. [Audience applauses] Um, I [Inaudible comments] ten minutes, yes exactly as planned, so.  
>> Hi, Hi
>> A decent player, yes managing time
>> Hi, here, here
>> Gonna ask question now?
>> Here, I have one. Um, so I recently saw a reddit post, about a composition that is Stockfish couldn’t solve. Eh, is it possible eh, to create a machine learning system that at the next [Inaudible] what problems are more likely to be solved for a human than any computer?
>> Did you hear the question?
>> It’s the, the sound is somehow, I don’t know why but the sound is just ah, it’s quite ah,
>> In your left, I mean in your left
>> It’s just again at the Def Con conference and he
>> You can ask your questions right here
>> Here, here
>> Oh, yeah, I can hear yes, okay
>> Alright, so I saw a composition that couldn’t be solved for a Stockfish, so
>> Yes, I mentioned Stockfish, but there are many other problems
>> Is possible to create a machine learning classifier that, eh, that takes what kind of position is, easier to play for human or more likely to be played better for human?
>> Um, look it’s the, it’s the first of all we don’t expect a machine to make a first move and to announce a mate in 17 555 moves, so, ah and um, I think definitely we can we can use machines just for, um, for the best recommendations for specific styles and that’s by the way what the top players are doing. They always looking for machines as the, sort of, as the, as the um, as the guides to um to help them to get to the positions that they like most. So, um because again, you have to just recognize that ah, machine evaluation is in, nine out of ten cases, is, is, far superior to, to the humans.
>> Alright, thanks
>> Hello, ah would you agree that real, hi
>> Yeah
>> Ah, would you agree that real intelligence, requires free will and free choices that only humans can make and Deep blue and any computer program is actually written by people and when you lose to Deep blue, you don’t lose to machine, you lose to a programmer, programmers of those programs.
>> Oh
>> So, my question is, do you think we are in any danger of, any kind of intelligence, until computer can have free will?
>> It’s ah, yeah, we are moving now from the, ah scientific domain to, to, to philosophy, ah. Ah, as for Deep blue, is very clear, it was a product of, of, of a great ward by, by humans. And I, you know I in, in most of the cases, we dealing even with AlphaGo and with Demis Hassabis team, it’s still the result of the work of, of, um, ah human intelligence. Now, um, whether machines can have a free will or not, I don’t know. Um, I used to; I used to believe that anything that we do while knowing how we do that, machines will do better. But, there are many things we do that without knowing how we do that, without even recognizing why it’s happening. And I don’t think it will be easy for machines, if possible at all, to grasp. So, for instance we have purpose, but we don’t know what purpose is. So that’s why I think it’s if you’re talking about free will which is somehow connected to the purpose. So, I think it’s ah; it might be very, very distant future for machines to ah get close to that.  
>> Thanks
>> Over here
>> Yeah
>> Um, what are your thoughts on human characteristics, such as bravery and morality and the decisions that artificial intelligence can make, related to the, ah for example the vehicle choosing to hit a child or go off a cliff and kill a driver?
>> Ah, that’s exactly the state you may call it passion because it’s the all different, you know, um human characteristics that cannot be quantified, at least easily quantified. And that’s the, that’s why I use the Hans Solo example because at the end of the day, when we talking about bravery it’s, it’s very often going against the odds. So, I think this, machines by definition will not be able to grasp it since, since they are basing, they, they, they based on, on, on sort of the best, finding the best patterns and so the best evaluations. And ah being brave and being passionate very often, in most of the cases goes against ah the, ah precise calculation.  
>> Mr Kasparov, I have a question a computer would not consider important. Ah, what’s in your flask and may I try some? [Laughter] What is contained in your flask?
>> I actually have the stoli that you pulled out of your pocket; I think that’s what he wants to know.
>> My pocket? Stolichnaya [Laughter] [Audience claps]
>> No that’s not an advertising [Laughter] you saw, you know, I just dropped it. [Laughter]  
>> Who, who will be the next human world champion? And do you think the young Chinese player; Wei Yi has a chance to dethrone Carlsen?
>> Um, currently Magnus Carlsen is, is the number one player, he’s not a world champion; he’s still the dominant player. Ah, he is 20; he’ll be turning 27 this year so he’s still quite young, though but not very young by the modern standards. I think Wei Li is 18 or 19, so um, I think Magnus will be facing younger players. There are two young American’s like ah Wesley So and Fabiano Caruana ah, and Wei Li, definitely, you know, by definition makes a potential challenger. Though it’s again, being a champion requires more than talent and being young and energetic. Ah, ah, you know, you need probably the element of luck, but Wei is definitely in the category of those who can and most likely will challenge Magnus Carlsen.
>> Thank you  
>> You discussed primarily the terministic algorithms or even basic machine learning when you were talking about using machines as tools to supplement our intelligence. However, ah, what do you say to the, immense amount of resources being important to, creating a strong AI or even ah, putting a human brain into a computer.  
>> Yeah, but um, again I always have to confess, you know, my ignorance and sometimes I’m just I’m not sure I’m in a position to answer a question, but something that always, you know, was I was struggling to understand what the human brains, let’s imagine you can just separated from our bodies, whether it can function separately, because I don’t know and this is probably, you guys know better. So how do brains functioning outside of the body, whether ,the, the, the  fact that it is moving so it is connected to our body, makes it, makes it work the way it works, maybe not. We don’t, we, we, again this is the kind of experiment that definitely will see maybe in, in, in the future. But my view is that, it’s the, it’s, it’s the combination of the movements and and other human factors and emotions, create, um the mind that is just, it’s bigger than just simply, you know just taking the brain and freezing it and using it as the, as the ah, um, device full of neurons.
>> Thank you
>> Yep
>> Hi, sorry, hi, um I was wondering ah, in light of the trend of machines eliminating human jobs, what are your thoughts on the idea of universal basic income?  
>> [Laughter] Yeah, it’s just this again the
>> Can you repeat the question?
>> Sorry okay ah, what are your thoughts on the idea of machines, oh sorry, of universal basic income?  
>> Um, no it’s the, I think it’s, it’s a very important question, because clearly we are moving at, moving ah, ah, reaching the point where a lot of people will be just be left behind. Since, um, it’s kind of a paradox of, of, um, of the technological progress. On one side, we have great new technologies, ah that make that, that that gives huge competitive advantage to younger people. Just, you know, every new generation is far more sophisticated, ah just by dealing with this, with this devices. On other side, we have the progress in medicine and the diet that, ah that helps people do to live longer and just to, to keep their, just you know ability to work for, for many, for longer years. So, but obviously, my generation, the fifties and of course sixties and even the forties, it’s just it’s, can, can hardly be competitive with, with the young kids just moving in. So, ah, we have to look for this, um, for this um, paradox and for, for this growing gap. Because we just have a gap that, that from history we know always led to big explosions. A gap between the social infrastructure of society and the technological ah, ah progress and ah, what you said is probably, it’s a part of the, part of the solution but the problem is that the politicians they are just trying to dump it, you know just to, to, to, to the next, to the next elections. Nobody wants to talk about it because it’s painful because it basically challenges the very foundation of the, of the sort of modern world order. It’s much easier to do [Inaudible] and keep printing money, so thinking that somebody else will pay. So there are many, many paradoxes that that make me feel uneasy because for instance, the piling debt will have to be paid by younger younger people. But will they be willing to do it and keeping the social guarantees for the older generation that made this debt? I think we are just, you know, [Audience claps] it’s, no it’s again I, I, I there are more questions that I,I can ask than, than, than, than answers that I can produce. Hopefully, I can help us with that. But at the end of the day, it’s, it’s, it’s very troubling that the political class are in the free world for ,for, for years, if not for decades, is trying to, to, to ignore the problems that we’re just discussing now. Because these problems will be, are already manifesting, they’re already on this plane and just, you know, ignoring the fact, you know, we have this, this technological progress, the, the huge development of, in many areas, is inevitably changing our lives. It’s, it’s extremely counterproductive and, and it’s basically neglecting our future. Thank you [Applause]
>> Thanks
>> That is all the time we have for questions, thank you very much everybody and thank you Mr Kasparov. [Applause]