Garry Kasparov: Chess, Deep Blue, AI, and Putin

[0] The following is a conversation with Gary Kasparov.

[1] He's considered by many to be the greatest chess player of all time.

[2] From 1986 until his retirement in 2005, he dominated the chess world, ranking world number one for most of those 19 years.

[3] While he has many historical matches against human chess players, in the long arc of history, he may be remembered for his match against the machine, IBM's Deep Blue.

[4] His initial victories and eventual loss to Deep Blue captivated the imagination of the world of what role artificial intelligence systems may play in our civilization's future.

[5] That excitement inspired an entire generation of AI researchers, including myself, to get into the field.

[6] Gary is also a pro -democracy political thinker and leader, a fearless human rights activist, and author of several books, including how life imitates chess, which is a book in strategy and decision -making, Winter is Coming, which is a book articulating his opposition to the Putin regime, and deep thinking, which is a book on the role of both artificial intelligence and human intelligence in defining our future.

[7] This is the Artificial Intelligence Podcast.

[8] If you enjoy it, subscribe on YouTube, give it five stars on iTunes, support it on Patreon, or simply connect with me on Twitter, Alex Friedman, spelled F -R -I -D -M -A -N.

[9] And now, here's my conversation with Gary Kasparov.

[10] As perhaps the greatest chess player of all time, when you look introspectively at your psychology throughout your career, what was the bigger motivator, the love of winning or the hatred of losing?

[11] Tough question.

[12] I have to confess, I never heard it before, which is, again, congratulations, it's quite an accomplishment.

[13] Losing was always painful.

[14] For me, it was almost like a physical pain because I knew that if I lost the game, it's just because I made a mistake.

[15] So I always believe that the result of the game had to be decided by the quality of my play.

[16] Okay, you may say it sounds arrogant, but it helped me to move forward because I always knew that there was room for improvement.

[17] Was there the fear of the mistake?

[18] Actually, fear of mistake guarantees mistakes.

[19] And the difference between top players and the very top is that it's the ability to make a decision without predictable consequences.

[20] You don't know what's happening.

[21] It's as intuitively.

[22] I can go this way or that way.

[23] And there are always hesitations.

[24] People are like, you are just, you know, at a crossroad.

[25] You can go right, you can go left, you can go straight.

[26] You can turn and go back.

[27] And the consequences are just very uncertain.

[28] You have certain ideas, what happens on the right or on the left or just, you know, if you go straight.

[29] But it's not enough to make well -calculated choice.

[30] And when you play chess at the very top, it's about your inner strength.

[31] So I can make this decision.

[32] I will stand firm and I'm not going to waste my time because I have full confidence that I will go through.

[33] Going back to your original question is, I would say neither.

[34] It's just it's the love for winning, hateful losing.

[35] There were important elements, psychological elements.

[36] But the key element, it's the, I would say the driving force was always my passion for making a difference.

[37] It's just I can move forward and I can always enjoy not just playing but creating something new.

[38] Creating something new.

[39] How do you think about that?

[40] It's just finding new ideas in the openings, you know, some original plan in the middle game.

[41] Actually, that helped me to make the transition from the game of chess where I was on the very top to another life.

[42] Where I knew I would not be number one.

[43] I would not be necessarily on the top.

[44] But I could still be very active and productive by my ability to make the difference by influencing people, say, joining the democratic movement in Russia or talking to people about human -machine relations.

[45] There's so many things where I knew my influence may not be as decisive as in chess, but still strong enough to help people to make their choices.

[46] So you can still create something new that makes a difference in the world outside of chess.

[47] But wait, you've kind of painted a beautiful picture of your motivations of chest, to create something new, to look for those moments of some brilliant new ideas.

[48] But were you haunted by something?

[49] You make it seem like to be at the level you're at, you can get away without having demons, without having fears.

[50] without being driven by some of the darker forces.

[51] I mean, you sound almost religious, the darker forces, spiritual demons.

[52] I mean, do you have a call for a priest?

[53] That's where I'm dressing.

[54] Now, just let's go back to these crucial chess moments where I had to make big decisions.

[55] As I said, it's, you know, it was all about my belief from very early days that I can make all the difference by playing well or by making mistakes.

[56] So, yes, I always had an opponent across the chessboard, opposite me. But no matter how strong the opponent was, whether they just were ordinary player or another world champion like Antony Karpov, having all respect for my opponent, I still believe that it's up to me to make the difference.

[57] And I, you know, I knew I was, not invincible.

[58] I made mistakes.

[59] I made some blunders.

[60] And, you know, with age, I made more blunders.

[61] Okay.

[62] I knew it.

[63] But it's still, you know, it's very much for me to be decisive factor in the game.

[64] I mean, even now, look, I just, you know, my latest chess experience was horrible.

[65] I mean, okay, I played Caruana, Fabi Caruana, this number two, number two, number three player in the world these days.

[66] We play this 960 with the Fisher, so -called Fisher random chess, reshuffling pieces.

[67] Yeah, I lost very badly, but it's because I made mistakes.

[68] I mean, I had so many winning positions.

[69] I mean, 15 years ago, I would have crushed him.

[70] So, and it's, you know, while I lost, I was not so much upset.

[71] I mean, I know, as I said in my interview, I can fight any opponent, but not my biological clock.

[72] So it's, it's fighting time is, is, is always a losing proposition.

[73] But even today, at age 56, you know, I, I knew that, you know, I could play great game.

[74] I couldn't finish it because I didn't have enough energy or just, you know, I couldn't have the same level of concentration.

[75] But, you know, in a number of games where I completely outplayed, one of the top players in the world, I mean, gave me a certain amount of pleasure.

[76] That is, even today, I haven't lost my touch.

[77] Not the same, you know, okay, the jaws are not as as strong and the teeth are not as sharp but I could get him just you know almost you know on the ropes still got it still got it and it's you know and it's I think it's my wife said it well I mean she said look Gary it's somehow it's only just fighting your biological clock it's just you know maybe it's a signal because you know the goddess of chess since you spoke great religiously the goddess of chess Keisha maybe she didn't want you to win because you know if you could beat number two, number three in the world.

[78] I mean, that's one of the top players who just recently played World Championship match.

[79] If you could beat him, that would really bad for the game of chess.

[80] People will say, oh, look, the game of chess, you know, it's not making any progress.

[81] The game is just, you know, it's totally devalued because look, the guy coming out of retirement, you know, just, you know, winning games.

[82] Maybe that was good for chess.

[83] Not good for you.

[84] But it's, look, I've been following your logic.

[85] we should always look for, you know, demons, you know, superior forces and other things that could, you know, if not dominate our lives, but somehow, you know, play a significant role in the outcome.

[86] Yeah, so the goddess of chess had to send a message.

[87] Yeah, that's okay, okay.

[88] So, Gary, you should do something else.

[89] Time.

[90] Now, for a question that you have heard before, but give me a chance.

[91] You've dominated the chess world for 20 years.

[92] and even still got it.

[93] Is there a moment you said you always look to create something new?

[94] Is there games or moments where you're especially proud of in terms of your brilliance of a new creative move?

[95] You've talked about Mikhail Tall as somebody who was aggressive and creative chess player in your own games.

[96] Look, you mentioned Michael Tall.

[97] It's very aggressive, very sharp player, famous for his combinations and sacrifices, even called magician from Riga, so for his very unique style.

[98] But any world champion, you know, it was a creator.

[99] Some of them were so flamboyant and flash -like tall.

[100] Some of them were no, just, you know, less discerned at the chess board like Tigrant.

[101] But every world champion, every top player brought something into the game of chess.

[102] And each contribution was priceless because it's not just about sacrifices.

[103] Of course, amateurs they enjoy, you know, the brilliant games where pieces are being sacrificed.

[104] It's all just, you know, it's all pieces of hanging.

[105] And it's all of a sudden, you know, being material down, a rook down or just, you know, queen down, the weaker side delivers the final blow on just, you know, mating opponents king.

[106] But there's other kinds of beauty.

[107] I mean, it's a slow position of maneuvering, you know, looking for weaknesses and just, and gradually, you know, strangling your opponent and eventually delivering sort of a positional masterpiece.

[108] So I think I made more difference in the game of chess than I could, I could have imagined when I started playing.

[109] And the reason I thought it was time for me to leave was that, I mean, I knew that I was not, I was not, not, no longer the position to, bring the same kind of contribution, the same kind of new knowledge into the game.

[110] And going back, I could immediately look at my games against Anzoli Karpov.

[111] It's not just I won the match in 1985 and became world champion at age 22, but there were at least two games in that match.

[112] Of course, the last one, game 24, that was decisive game of the match.

[113] I won and became world champion.

[114] But also the way I won it was a very sharp game and I found a unique maneuver that was absolutely new and it became some sort of just a typical now.

[115] Though this when the move was made on the board and put on display, a lot of people thought it was ugly.

[116] And another game, game 16 in the match where I just also managed to outplay carp of completely with black pieces, just paralyzing his entire army in its own camp.

[117] Technically or psychologically, or is that a mix of both in Game 16?

[118] Yeah, I think it was a big blow to Carpuff.

[119] I think it was a big psychological victory for number of reasons.

[120] One, the score was equal at a time.

[121] And the world champion, you know, by the rules, could retain his title in case of a tie.

[122] So we still have, you know, before Game 16, we have nine games to go.

[123] And also it was some sort of a bluff, because neither mean or Karpov saw the refutation of this opening idea.

[124] And I think it says for Karpo, it was double blow, because not that he lost the game, I should triple blow.

[125] He lost the game.

[126] It was a brilliant game.

[127] And I played impeccably after, you know, just this opening bluff.

[128] And then, you know, they discovered that it was a bluff.

[129] So it's the, again, I didn't know.

[130] I was not bluffing.

[131] So that's why it happens very often.

[132] It's when, you know, some ideas could be refuted.

[133] And it's just what I found out.

[134] And this is, again, going back to your spiritual theme, is that you could spend a lot of time working.

[135] And when I say you could, it's in the 80s, in the 90s.

[136] It doesn't happen these days because everybody has a computer.

[137] You could immediately see if it works or it doesn't work.

[138] Machine shows your refutation in a split of a second.

[139] But many of the analysis in the 80s or in the 90s, they were not perfect simply because we were humans and you analyze the game, you look for some fresh ideas.

[140] And then just it happens that there was something that you missed because the level of a concentration at the chess board is different from one that when you analyze the game, just moving the pieces around.

[141] But somehow, if you spend a lot of time at the chess board preparing, so in your studies, with your coaches, hours and hours and hours, and nothing of what you found could, you know, had materialized on the, on the, on the chess board.

[142] Somehow these hours helped, I don't know why, always helped you.

[143] It's as if, you know, the amount of work you did could be transformed into some sort of spiritual energy that helped you to come up with other great ideas during the board.

[144] Again, even if there was no direct connection between your preparation and your victory in the game, there was always some sort of invisible connection between the amount of work you did, your dedication to actually and your passion to discover new ideas, and your ability during the game at the chessboard when the clock was ticking, we still had ticking clock, not digital clock at the time.

[145] So to come up with some brilliance.

[146] And I also can mention many games from the 90s, so it's the, obviously all amateurs would people, up my game against Vesely and Topoloff in 1999 and V. Gonzay, again, because it was a brilliant game, the Black King traveled from its own camp to into the White's camp across the entire board.

[147] It doesn't happen often, trust me, as you know, in the games with professional players, top professional players.

[148] So that's why visually it was one of the most impressive victories.

[149] But I could bring to our attention.

[150] Many other games that were not so impressive for, for amateurs, not so, not so beautiful, just because it's sacrifice is always beautiful.

[151] You sacrifice, and then, and then eventually you have so, you have very few resources left, and you, you, you use them just to, to, to, to crush your, your, your opponent, basically, to, it's, you have to make the king because you have almost, almost, nothing, nothing, nothing left at your disposal.

[152] But I, you know, up to the very end, get less and less, but still up to the very end, I always had games with some sort of, you know, interesting ideas and games that gave me great satisfaction.

[153] But I think it's what happened from 2005 up to these days was also a very big accomplishment since, you know, I had to find myself to sort of relocate myself.

[154] Yeah, re -channel the creative energies.

[155] Exactly.

[156] And to find something where I feel comfortable, even confident that my participation still makes the difference.

[157] Beautifully put.

[158] So let me ask perhaps a silly question, but sticking on chest for just a little longer, where do you put Magnus Carlson, the current world champion in the list of all -time grades?

[159] In terms of style, moments of brilliance, consistency.

[160] It's a tricky question.

[161] You know, the moment you start ranking, Yeah, you lose something.

[162] It's the, I think it's, it's not fair because it's the, any new generation knows much more about the game than the previous one.

[163] So when people say, oh, Gary was the greatest, Fisher was the greatest, Magnus was the greatest, it disregarded the fact that the great players of the past, the last year, Capablanca, Alokian, I mean, they knew so little about chess.

[164] By today's standards.

[165] I mean, today, just any kid, you know, that spent a few years, you know, with his or her chess computer knows much more about the games, simply just because you have access to this information.

[166] And it has been discovered, generation after generation.

[167] We added more and more knowledge to the game of chess.

[168] It's about the gap between the world champion and the rest of the field.

[169] So it's the...

[170] Now, if you look at the gap, then probably Fisher, you know, could be on top.

[171] but very short period of time.

[172] Then you should also add a time factor.

[173] I was on top not as big as Fisher, but much longer.

[174] So, and also, unlike Fisher, I succeeded in beating next generation.

[175] Here's the question.

[176] Let's see if you still got the fire.

[177] Speaking of the next generation, because you did succeed beating the next generation.

[178] Next, it's close.

[179] Okay, Annand, Short, Annand, the Shire of.

[180] Cromney is already 12 years younger, so that's a next.

[181] But still, yet, I competed with that.

[182] and I just, I beat most of them, and I was still dominant when I left at age of 41.

[183] So back to Magnus.

[184] Magnus, I mean, consistency is phenomenal.

[185] The reason Magnus is on top and it seems unbeatable today, Magnus is a lethal combination of Fisher and Karpov, which is very, it's very unusual because Fisher's style is very dynamic, just fighting to the last poem, just using every reason.

[186] resource available.

[187] Carbon was very different.

[188] It's just an unparalleled ability to use the every piece with a maximum effect.

[189] Just it's minimal resources always produce maximum effect.

[190] So now imagine that you merge these two styles.

[191] So it's it's like, you know, it's squeezing every stone for a drop of water.

[192] But but doing it, you know, just, you know, for 50, 60, 70, 70, any moves.

[193] I mean, Magnus could go on as long as Fisher with all his passion and energy, and at the same time being as meticulous and deadly as carpal by just, you know, using every little advantage.

[194] So, and it has good, you know, very good health.

[195] It's important.

[196] I mean, physical conditions are, by the way, very important.

[197] So a lot of people don't recognize it.

[198] There are latest studies shows that chess players burned thousands of calories during the game.

[199] So that puts him on the top of this field of the world champions.

[200] But again, it's the discussion that is, I saw recently in the internet, whether Gary Kasparov of his peak, let's say, late 80s could beat Magnus Carlson today.

[201] I mean, it's totally irrelevant because Gary Kasparov in 1989, okay, it's played great chess, but still I knew very little about chess compared to Magnus Carlson in 2019, who, by the way, learned from me as well.

[202] So that's why, yeah, I'm extremely cautious in making any judgment that involves, you know, time gaps.

[203] You ask, you know, soccer fans, so who is your favorite, Pelle, Maradona or Messi?

[204] Yeah.

[205] Yeah, who's your favorite?

[206] Messi.

[207] Messy.

[208] Yeah, why?

[209] Maybe Maradona, maybe.

[210] No, because you're younger, but that's simple.

[211] Your instinct of answer is correct because you saw, you didn't say Maradona in action.

[212] I saw all of them in action, so that's why.

[213] But since, you know, when I was.

[214] you know, just following it, you know, just, it's Pele and Maradona.

[215] They were just, you know, there were big stars.

[216] And it's, Messies already just, I was gradually losing interest, just other things.

[217] So I remember Pele in 1970, the final match Brazil, Italy.

[218] So that's the first World World Cup soccer I watched.

[219] So that's the, and actually my answer when I just, you know, because I was asked this question as well.

[220] So I say that it's just while it's impossible to make a choice, I would still probably go with Maradona for a simple reason.

[221] the Brazilian team in 1970 could have won without Pellet.

[222] It was absolutely great.

[223] Still could have won.

[224] Maybe, but it is, the Argentinian team in 1986 without Maradona would not be in the final.

[225] So this is, and Messi, he still hasn't won a title.

[226] You could argue for that for an hour, but you could say if you ask Maradona, if you look in his eyes, especially, let's say, Gary Kasparov, in 1989, he would have said, I was sure as hell would beat Magnus Carl.

[227] And just simply because simply because, again, it's this, they saw me in action.

[228] So this, again, it's the age factor as important.

[229] Therefore, with the passion and energy and being equipped with all modern ideas.

[230] But again, then you make, you know, a very just important assumption that you could empower Gary Kasparo for 89 with all ideas that have been accumulated over 30 years.

[231] That would not be Gary Kasparra.

[232] That would be someone else.

[233] Yeah.

[234] Because, again, I belong to 1989.

[235] I was way ahead of the field, and I, you know, I beat carp off several times in a world championship matches, and I crossed 2 ,800, which, by the way, if you look at the in rating, which is just, it's, even today, so this is, this is the rating that I retire.

[236] So that's still, you know, it's just, it's a top two, two, three.

[237] So that's this, it's Caruana and D. It's about the same rating now.

[238] And I crossed 2 ,800 in 1990.

[239] We just, you look at the inflation.

[240] When I crossed 2 ,800, in 1990, there was only one player in 2700 category, Anatoly Carpuff.

[241] Now he had more than 50.

[242] So just when you see this, so if you add inflation, so I think my 2851, it probably could be more valuable as Magnus 2882, which was his highest rating.

[243] But anyway, again, too many hypotheticals.

[244] You're lost to IBM D. Blue in 1997.

[245] In my eyes, that is one of the most seminal moments in the history.

[246] Again, I apologize.

[247] for being romanticizing the notion, but in the history of our civilization, because humans as a civilization for centuries saw chess as the peak of what man can accomplish, of intellectual mastery, right?

[248] And that moment when a machine could beat a human being was inspiring to just an entire, anyone who cares about science, innovation, an entire generation of AI researchers, And yet, to you, that loss, at least if reading your face, was seemed like a tragedy, extremely painful.

[249] Like you said, physically painful.

[250] Why?

[251] When you look back at your psychology of that loss, why was it so painful?

[252] Were you not able to see the seminal nature of that moment?

[253] Or was that exactly why was that painful?

[254] As I already said, losing was painful.

[255] physically paid.

[256] And the match I lost in 1997 was not the first match I lost to a machine.

[257] It was the first match I lost, period.

[258] Yeah.

[259] That's...

[260] Oh, wow.

[261] Oh, wow.

[262] Yeah, it's...

[263] Right.

[264] Yeah.

[265] That makes all the difference to me. Yes.

[266] First time I lost.

[267] It's just...

[268] Now, I lost, and the reason I was so angry that I just, you know, I had...

[269] suspicions that my loss was not just the result of my bad play.

[270] So though I played quite poorly, you know, just when you started looking at the games today, I made tons of mistakes, but, you know, I had all reasons to believe that, you know, there were other factors that had nothing to do with the game of chess.

[271] And that's why I was angry.

[272] But look, it was 22 years ago.

[273] It's water under the bridge.

[274] We can analyze this match and this is with everything you said.

[275] I agree with probably one exception is that.

[276] Considering chess, you know, as the sort of as a pinnacle of intellectual activities was our mistake.

[277] Because, you know, we just thought, oh, it's a game of the highest intellect and it's just, you know, you have to be so, you know, intelligent.

[278] And you could see things that, you know, the ordinary mortals could not see.

[279] It's a game.

[280] And all machines had to do in this game is just to make fewer mistakes, not to solve the game.

[281] because the game cannot be solved.

[282] I mean, according to Colchanan, the number of legal moves is 10 to the 46 power.

[283] Too many zeros.

[284] So just for any computer to finish the job, you know, in next few billion years.

[285] But it doesn't have to.

[286] It's all about making fewer mistakes.

[287] And I think that's this match, actually, and what's happened afterwards with other games, with Goh, with Shoggy, with video games.

[288] It's a demonstration that machines will always be humans in what I call closed systems.

[289] The moment you build a closed system, no matter how this system is called, chess, go, shoggy, daughter, machines will prevail simply because they will bring down a number of mistakes.

[290] Machines don't have to solve it.

[291] They just have to, the way they outplay us, it's not by just being more intelligent.

[292] It's just by doing something else, but eventually it's just, it's capitalizing on our mistakes.

[293] When you look at the chess machines today, compare this to Magnus Carlson, it's the same as comparing Ferrari to Hussein Bolt.

[294] The gap is, I mean, by chess standards, is insane.

[295] 34, 3 ,500 to 2800, 2850 on Magnus.

[296] It's like difference between Magnus and an ordinary player from an open.

[297] an international tournament.

[298] It's not because machine understand is better than Magnus Carlson, but simply because it's steady.

[299] Machine has steady hand.

[300] And I think that is what we have to learn from 997 experience and from further encounters with computers and sort of the current state of affairs with Alpha Zero, you were beating other machines.

[301] The idea that we can compete with computers in so -called intellectual fields.

[302] It's, it was wrong from the very beginning.

[303] It's just, by the way, the 997 match was not the first victory of machines over grandmasters.

[304] Over grandmasters.

[305] Yeah.

[306] Now, actually, it's, I played against first decent chess computers from late, from late 80s.

[307] So I played with the prototype of Deep Blue called Deep Thought in 1989, two rapid chess games in New York.

[308] I won handily, both games, we played against new chess engines like Fritz and other programs and then it was Israeli program Junior that appeared in 1995.

[309] Right, right, right, I remember.

[310] Yeah, so there were several programs.

[311] I, you know, I lost few games in Blitz.

[312] I lost one match against the computer, a chess engine in 1994, rapid chess.

[313] So I lost one game to the blue in 1996 match, the match I won.

[314] Some people, you know, tend to forget about it that I won the first match.

[315] Yes.

[316] But it's, it's, we, we made a very important psychological mistake thinking that the reason we lost blitz matches, five, five minutes games, the reason we lost some of the rapid chess matches, 25 minutes chess, because we didn't have enough time.

[317] If you play a longer match, we will not make the same mistake.

[318] Nonsense.

[319] So, yeah, we had more time, but we still make mistakes.

[320] And machine also has more time.

[321] And machines, machine will always, you know, we always, you know, we always.

[322] be stated inconsistent compared to humans' instabilities and inconsistencies.

[323] And today we are at a point where, yes, nobody talks about, you know, humans playing as machines.

[324] Now machines can offer handicap to top players and still, you know, will be favored.

[325] I think we're just learning that it's no longer human versus machines.

[326] It's about human working with machines.

[327] That's what I recognized in 1998 Just after leaking my wounds and spending one year and just, you know, ruminating so the, so what's happened in this match.

[328] And I knew that though we still could play against the machines.

[329] I had two more matches in 2003 playing both Deep Fritz and Deep Junior.

[330] Both matches ended as a tie.

[331] Though these machines were not weaker at least, actually probably stronger than Deep Blue.

[332] And by the way, today, chess app on your mobile phone is probably stronger than Deep Blue.

[333] I'm not speaking about chess engines that are so much superior.

[334] And by the way, when you analyze games we played against the Blue 997 on your chess engine, they'll be laughing.

[335] And it also shows that how chess changed because chess commentators they'll look at some of our games like game four, game five, brilliant idea.

[336] Now you ask stockfish, you ask Houdini, you ask Commodore, all the leading chess engines.

[337] Within 30 seconds, they will show you how many mistakes.

[338] Bose, Gary and D. Blue made in the game that was trumpeted as a great chess match in 1997.

[339] Well, okay, so you've made an interesting, if you can untangle that comment.

[340] So now in retrospect, it was a mistake to see chess as the peak of human intellect.

[341] Nevertheless, that was done for centuries.

[342] By the way, in Europe, because.

[343] You know, you move to the Far East.

[344] They will go.

[345] But games, games.

[346] Again, some of the games, like, you know, are board games.

[347] Yes.

[348] Yes.

[349] Yeah, I agree.

[350] So if I push back a little bit.

[351] So now you say that, okay, but it was a mistake to see chess as the epitome.

[352] And now, and then now there's other things maybe, like language, that conversation, like some of the things that in your view is still way out of reach of computers, but inside humans.

[353] Can you talk about what those things might be and do you think just like chess that might fall?

[354] With the same set of approaches if you look at Alpha Zero, the same kind of learning approaches as the machines grow in size.

[355] No, it's not about growing in size.

[356] It's about, again, it's about understanding the difference between closed system and open -ended system.

[357] So you think that key difference, so the board games are closed in terms of the rule set, the actions, the state space, everything is just constrained.

[358] You think once you open it, the machines are lost?

[359] Not lost, but again, the effectiveness is very different because machine does not understand the moment.

[360] It's reaching territory of diminishing returns.

[361] It's the, to put it in a different way, machine doesn't know how to ask right questions.

[362] It can ask questions, but it will never tell you which questions are relevant.

[363] It's the, it's like about the, it's a direction.

[364] So these, it's, I think it's in human -machine relations, we have to consider so our role.

[365] And people, many people feel uncomfortable that this, the territory that, that belongs to us is, is shrinking.

[366] I'm saying, so what, you know, this is, eventually we'll belong to the last few decimal points.

[367] But it's like having, so very powerful gun.

[368] And, and, and all you can do there is slightly.

[369] alter direction of the bullet, maybe 0 .1 degree of this angle.

[370] But that means a mile away, 10 meters of the target.

[371] So we have to recognize that it's a certain unique human qualities that machines in a foreseeable future will not be able to reproduce.

[372] And the effectiveness of this cooperation, collaboration, depends on our understanding what exactly we can bring into the game.

[373] So the greatest danger is when we try to interfere with machine superior knowledge.

[374] So that's why I always say that sometimes you'd rather have, by reading these pictures in radiology, you may probably prefer an experience nurse rather than having a top professor because she will not try to interfere with machines' understanding.

[375] So it's very important to know that if machines knows how to do better things in 95%, 96 % of territory, we should not touch it because it's it happened.

[376] It's like in chess.

[377] Recognize.

[378] They do it better.

[379] See where we can make the difference.

[380] You mentioned alpha zero.

[381] I mean, alpha zero is it's actually a first step into what you may call AI because everything that's being called AI today is just, it's it's one or another variation of what Claude Shannon characterized as a brute force.

[382] It's a type A machine, whether it's D -Blue, whether it's what's and all these things.

[383] It's the modern technologies that are being trumpeted as AI.

[384] It's still brute force.

[385] All they do is they do optimization.

[386] It's this.

[387] They are, you know, they keep, you know, improving the way to process human -generated data.

[388] Now, Alpha -Zero is the first step towards, you know, machine -produced knowledge.

[389] Yes.

[390] Which is, by the way, it's quite ironic that the first company that championed that, was IBM.

[391] Oh, it's in baggammon.

[392] Interesting.

[393] In backgammon.

[394] Yes, you just, you should, you should, you should look at IBM.

[395] It's a neurogammon.

[396] It's the, it's the scientist process.

[397] He's still working at IBM.

[398] They had in early 90s.

[399] It's just, it's the, it's the program that played, you know, the alpha zero type.

[400] So just trying to come up with own strategies.

[401] But because of success of the Blue, this project had been not abandoned, but just, you know, it was put on call.

[402] And now we just, you know, it's, it's, you know, it's, you know, it's everybody talks about, about this, the, the machines generated knowledge.

[403] So as revolutionary.

[404] And it is.

[405] But there's still, you know, many open -ended questions.

[406] Yes, Alpha -Zero generates its own data.

[407] Many ideas that Alpha -0 generated in chess were quite intriguing.

[408] So I, I looked at these games with, not just with interest, but with, you know, it was quite exciting.

[409] to learn how machine could actually, you know, juggle all the pieces and just play positions with a broken material balance, sacrificing material, always being ahead of other programs, you know, one or two moves ahead by foreseeing the consequences, not over -calculating, because other machines were at least as powerful in calculating.

[410] But it's having this unique knowledge based on discovered patterns after playing 60 million games.

[411] Almost something that feels like intuition.

[412] Exactly.

[413] But there's one problem.

[414] Now, the simple question, if Alpha Zero faces superior a point, let's say another powerful computer accompanied by a human who could help just to discover certain problems.

[415] Because I look at many Alpha Zero games.

[416] I visited their lab, spoke to Demis Khazabis and his team.

[417] I know there's certain witnesses there.

[418] Now, if this witness are exposed, the question is, how many games will it take for Alpha Zero to correct it?

[419] The answer is hundreds of thousands.

[420] Even if it keeps losing, it's just because the whole system is based.

[421] So it's now, imagine so this as you can have a human by just make a few tweaks.

[422] So humans are still more flexible.

[423] And as long as we recognize what is our role, where we can play sort of, so the most valuable part in this collaboration.

[424] So it will help us to understand what are the next steps in human machine collaboration.

[425] beautifully put so let's talk about the thing that machines certainly don't know how to do yet which is morality machines and morality it's another question that you know just it's being asked all the time these days and I think it's another phantom that is haunting a general public because it's just being fed with this you know illusions is that how can we avoid machines you know having bias being prejudices you cannot because it's like looking the mirror and complaining about what you see.

[426] If you have certain bias in the society, machine will just follow it.

[427] It's just, it's, you know, you look at the mirror, you don't like what you see there.

[428] You can, you know, you can break it, you can try to distort it.

[429] Or you can try to actually change something just by yourself.

[430] By yourself, yes.

[431] So it's very important to understand.

[432] You cannot expect machines to improve the yields of our society.

[433] And moreover, machines will.

[434] simply, you know, just, you know, amplify.

[435] Yes.

[436] But the thing is, people are more comfortable with other people doing injustice with being biased.

[437] We're not comfortable with machines having the same kind of bias.

[438] So that's an interesting standard that we place on machines with autonomous vehicles.

[439] They have to be much safer with automated systems.

[440] Of course, they're much safer.

[441] Statistically, they're much safer than that.

[442] It's not, of course.

[443] Why would it?

[444] It's not of course.

[445] It's not given.

[446] Autonomous vehicles, you have to work really hard to make them safer.

[447] I think it just goes without saying is the outcome of the, of this, I would call a competition with comparison is very clear.

[448] But the problem is not about being, you know, safer.

[449] It's the 40 ,000 people or so every year died in car accidents in the United States.

[450] and it's statistics.

[451] One accident with autonomous vehicle and its front page with newspaper.

[452] Yes.

[453] So it's, again, it's about cycle.

[454] So it's while people, you know, kill each other in car accidents because they make mistakes, they make more mistakes.

[455] For me, it's not a question.

[456] Of course we make more mistakes because we're human.

[457] Yes, machines, and by the way, no machine will ever reach 100 % perfection.

[458] That's another important fake story that is being fed to the public.

[459] If machine doesn't reach 100 % performance is not safe.

[460] No. All you can ask any computer, whether it's, you know, playing chess or doing the stock market calculations or driving your autonomous vehicle, it's to make fewer mistakes.

[461] And yes, I know it's not, you know, it's not easy for us to accept because, ah, if, you know, if you have two humans, you know, colliding in their cars, okay, it's like, if one of one of these cars is autonomous vehicle.

[462] And by the way, even if it's humans fault, terrible.

[463] How How could you allow a machine to run without a driver at the wheel?

[464] So, you know, let's linger that for a second, that double standard.

[465] The way you felt with your first loss against D. Blue, were you treating the machine differently than you would have a human?

[466] So what do you think about that difference between the way we see machines and humans?

[467] No, at that time, you know, for me, it was a match.

[468] And that's why I was angry because I believed that the loss.

[469] The match was not, you know, fairly organized.

[470] So it's definitely there were unfair advantages for IBM and I wanted to play another match, like a rubber match.

[471] So your angered or displeasure was aimed more like at the humans behind IBM versus the actual pure algorithm.

[472] Absolutely.

[473] Look, I knew at a time, and by the way, I was objectively speaking, I was stronger at that time.

[474] So that's probably added to my anger because I knew I could beat machine.

[475] Yeah.

[476] Yeah.

[477] So this, and that's the, and I lost, and I knew I was not well prepared, so because they, I have to give them credit.

[478] They did some good work from 1996, and I, but I still could beat the machine.

[479] So I made too many mistakes.

[480] Also, this is the whole, it's this, the publicity around the match.

[481] So I, I underestimated the effect, you know, just it's, and the, and being called, you know, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, no, no pressure.

[482] Okay.

[483] Okay, well, let me ask.

[484] So I was born also in the Soviet Union.

[485] What lessons do you draw from the rise and fall of the Soviet Union in the 20th century?

[486] When you just look at this nation that is now pushing forward into what Russia is, if you look at the long arc of history of the 20th century, what do we take away?

[487] What do we take away from that?

[488] I think the lesson of history is clear, undemocratic systems, totalitarian regimes, systems that are based on controlling their citizens and just every aspect of their life, not offering opportunities to, for private initiative, central planning systems, they doomed.

[489] They just, you know, they, they cannot be driving force for innovative.

[490] So they, in a history timeline, I mean, they could cause certain, you know, distortion of, of, the concept of progress.

[491] They, by the way, they may call themselves progressive, but we know that the damage that they cost to humanity is just, it's yet to be measured.

[492] But at the end of the day, they fail.

[493] They fail and it's, and the end of the Cold War was a great triumph of the free world.

[494] It's not that free world is perfect.

[495] It's very important to recognize the fact that I always like to mention, you know, one of my favorite books, The Lord of the Rings, that there's no, there's no absolute good, but there is an absolute evil.

[496] Good, you know, comes in many forms, but we all, you know, it's being humans or being even, you know, humans from fairy tales or just some sort of mythical creatures.

[497] It's the, you can always find spots on the sun.

[498] So this is, you can, you can.

[499] conducting war and fighting for justice, there are always things that, you know, can be easily criticized.

[500] And human history is a never -ending quest for perfection.

[501] But we know that there's absolutely evil.

[502] We know it's, for me, it's no clear.

[503] I mean, nobody argues about Hitler being absolutely evil, but I think it's very important to reckon that Stalin was absolutely evil.

[504] Communism caused more damage than any other ideology in the 20th century.

[505] And unfortunately, while we all know that fascist was condemned, but there was no Nuremberg for communism.

[506] And that's why we could see, you know, still the successors of Stalin are feeling far more comfortable.

[507] So you, you, is one of them.

[508] You highlight a few interesting connections, actually, between Stalin and Hitler.

[509] I mean, in terms of the adjusting or clarifying the history of World War II, which is very interesting.

[510] Of course, we don't have time.

[511] So let me ask.

[512] As you can ask, I just recently delivered a speech in Toronto at 8th anniversary of Moritz of Ribbon -Trump Act.

[513] It's something that I believe, you know, just, you know, must be taught in the schools that the World War II had been started by two dictators, by signing these criminal treaty, collusion of two tyrants in August 1939 that led to the beginning of the World War II.

[514] And the fact is that eventually Stalin had no choice but to join allies.

[515] because Hitler attacked him, so it just doesn't, you know, eliminate the fact that Stalin helped Hitler to start World War II.

[516] And he was one of the beneficiaries at early stage by annexing part of Eastern Europe.

[517] And as a result of the World War II, he annexed almost entire Eastern Europe.

[518] And for many Eastern European nations, the end of the World War II was the beginning of communist occupation.

[519] So Putin, you've talked about as a man who stands between Russia and democracy, essentially today.

[520] You've been a strong opponent and critic of Putin.

[521] Let me ask again, how much does fear enter your mind and heart?

[522] So in 2007, there's this interesting comment from Oleg Kalugan, KGB General.

[523] He said that I do not talk details.

[524] People who knew them are all dead now because they were vocal.

[525] I'm quiet.

[526] there's only one man who is vocal and he may be in trouble world chess champion casparov he has been very outspoken in his attacks on Putin and i believe he's probably next on the list so clearly your life has been and perhaps continues to be in danger how do you think about having the views you have the ideas you have being in opposition as you are in this kind of context when your life could be in danger.

[527] That's the reason I live in New York.

[528] So it was not my first choice, but I knew I had to live Russia at one point.

[529] And among other places, New York is the safest.

[530] Is it safe?

[531] No. I mean, it's the, I know what happened, what is happening with many of Putin enemies.

[532] But at the end of the day, I mean, what can I do?

[533] I could be very proactive by trying to change things.

[534] I can influence.

[535] But here are we a fact.

[536] I cannot stop doing what I've been doing for a long time.

[537] It's the right thing to do.

[538] I grew up with my family teaching me, sort of the wisdom of Soviet dissidents, do what you must and so be.

[539] I could try to be cautious by not traveling to certain places where my security could be at risk.

[540] There's so many invitations to speak at different locations in the world.

[541] I have to say that many countries are just now are not destinations that I can afford to travel.

[542] My mother still lives in Moscow.

[543] I meet her a few times a year.

[544] She was devastated when I had to leave Russia because since my father died in 1971, so she was 33 and she dedicated her entire life to her only son.

[545] But she recognized in just a year or so since I left Russia that it was the only chance, for me to continue my normal life.

[546] So just to, I mean, to be relatively safe and to do what she taught me to do, to make the difference.

[547] Do you think you will have a return to Russia?

[548] Or let me ask a different way.

[549] Even sooner than many people think because I think Putin regime is facing unsurmountable difficulties.

[550] And again, I read enough historical books to know that dictatorships, they end suddenly.

[551] It's just, on Sunday dictator feels comfortable.

[552] He believes he's popular on Monday morning.

[553] He's bust.

[554] The good news and bad news.

[555] I mean, the bad news is that I don't know when and how Putin rule ends.

[556] The good news, he also doesn't know.

[557] Okay, well put.

[558] Let me ask a question.

[559] that seems to preoccupy the American mind from the perspective of Russia.

[560] One, did Russia interfere in the 2016 U .S. election, government sanction, and future, two, will Russia interfere in the 2020 U .S. election?

[561] And what does that interference look like?

[562] It's very old, you know, we had such an intelligent conversation.

[563] And you are ruining everything by asking such a stupid question.

[564] It's been going downhill the entire way.

[565] Yeah, but it's insulting for my intellect.

[566] Of course they did interfere.

[567] Of course they did absolutely everything to elect Trump.

[568] I mean, they said it many times.

[569] It is just, you know, I met enough KGB colonels in my life to tell you that, you know, just the way Putin looks at Trump.

[570] This is the way looks.

[571] And I don't have to hear what he says, what Trump says, you know, just says, I don't need to go through congressional investigations.

[572] The way Putin looks at Trump is the way the KGB officers looked at their assets.

[573] It's just, and falling to 2020, of course they will do absolutely everything to help Trump to survive because I think the damage that Trump's re -elections could cause to America and to the free world, it's just, it's beyond one's imagination.

[574] I think basically if Trump is re -elected, he will ruin NATO because he's already heading in this direction, but now he's just, he's still limited by the re -election.

[575] hurdles if he's still in the office after November 2020 January 2021 I don't want to think about it my problem is not just Trump because Trump is basically a symptom but the problem is that I don't see it's the in American political horizon politicians who could take on Trump for all damage that he's doing for the free world, not just things that that's happened that went wrong in America.

[576] So it seems to me that the campaign, political campaign on the Democratic side is fixed on certain important, but still secondary issues.

[577] Because when you have the foundation of the Republican jeopardy, you cannot talk about health care.

[578] I mean, understand how important it is, but it's still secondary because the entire framework of American political life is at risk.

[579] And you have Vladimir Putin just, you know, just, it's having, fortunately, free hands by, by his, by attacking America and other free countries.

[580] And by the way, we have so much evidence about Russia interference in Brexit, in elections in almost every European country.

[581] And thinking that they will be shy of attacking America in 2020, now with Trump in the office, yeah, I think it's, yeah, it definitely diminishes the intellectual.

[582] intellectual quality falconvers.

[583] I do what I can.

[584] Last question.

[585] If you can go back, just look at the entirety of your life.

[586] You're accomplished more than most humans will ever do.

[587] If you can go back and relive a single moment in your life, what would that moment be?

[588] There are moments in my life when I think about what could be done differently.

[589] but no experience happiness and joy and pride just just to just to touch once again I know I know I made many mistakes in my life so I just it's the I know with that at the end of the day it's I believe in the butterfly effect so it's the I knew moments where I could now if I'm there at that point in 89 in 93 pick up a year I could improve my actions by not doing this stupid thing.

[590] But then how do you know that I will have all other accomplishments?

[591] Yeah, I just I'm afraid that, you know, we just have to just follow this, if you may call wisdom of forest gump, you know?

[592] It's the life is this, you know, it's a box of chocolate and you don't know what's inside, but you have to go one by one.

[593] So it's the, I'm happy with who I am and where I am today.

[594] And I am very proud, not only with my chess accomplishments, but that I made this transition.

[595] And since I left chess, you know, I built my own reputation that had some influence of the game of chess, but not, it's not, you know, directly derived from the game.

[596] I'm grateful for my wife, so help me to build this life.

[597] We actually married in 2005.

[598] It was my certain marriage.

[599] That's why I said I made mistakes in my life.

[600] But, and by the way, I'm close with two kids from my previous marriages.

[601] So that's the, I'm, you know, I managed to sort of to balance my life.

[602] And here, you know, I live in New York, so we have two kids born here in New York.

[603] It's new life and it's, you know, it's busy.

[604] Sometimes I wish I could, you know, I could limit my engagement in many other things that are still, you know, taking time and energy.

[605] But life is exciting.

[606] And as long as I can feel that I have energy, I have, strengths, I have passion to make the difference.

[607] I'm happy.

[608] I think that's a beautiful moment to end on Gary.

[609] Thank you very much for talking today.

[610] Thank you.

[611] Thank you.