[THE DIGITAL FACTORY] Nassim Nicholas Taleb on Antifragile

thank you i'm extremely honored to be here and i would like to thank my friends for printing and discovering the practical use of a 3d printer with a black swan being printed by my friends here however i'll make a comment i think it'd be a little more difficult to print something called uncertainty the title of the inserto collection so one but i'm certain they'll find a way to print something called uncertainty once you can visualize it but i guess we're a few centuries away from visualizing something called uncertainty so i'm introducing myself here i was my origin as trader because a lot of people think i'm a professor a lot of people think i'm a writer so i give a little bit of so that's my national origin i was a traitor doing these kind of weird things and then i got downgraded downgraded myself to becoming a pit trader those people who you know stand in a bit sweat a lot you know the smell is not conveyed by the picture but you can understand and you spent six hours on your feet i did that from qualitative trader you know for the experience and i fell in love with marcus how prices um how prices develop price formation is very difficult uh saying still don't understand it then it was so difficult trading in a pit that i went to do this for retirement largely because i can't do tennis so this was easier so i became a professor at nyu and focusing on things so it's pretty much backwards so i went from practice sorry this doesn't go backwards no this doesn't go across yeah it does so i went from practice to theory uh rather than go from theory to practice which means that you make enemies in theory and you make enemies in practice because you know you don't have the same culture in both and now this is the technical insert though i'm embarking on and one of the first volume is out it's on something we'll discuss in in in the uh during the the fire uh the fire uh chat part success but the the first volume is about so uh fat tails how most of history comes from [Music] um random events that are very consequential and what an environment that i call scalable and i've discovered yesterday visiting my friends that this becomes scalable whatever it was was uh limited to artisanal production become scalable i mean surgery is not scalable yet the surgeon the best surgeon in the world cannot see a million people in a day but an author you write a book once and that's it luckily you can get i'm very lazy so it's great to write a book every time someone buys it you don't have to rewrite it you see unlike surgery or other professions but uh thanks to this a lot of professors are becoming scalable you know even dentistry i mean you can make a million teeth all right you don't have to be there except for the first one anyway so this is that was the first uh block of research fat tales what i call fat tails extreme events on which later now let's talk about fragility so before we talk about robustness resilience stuff like that we need to define fragility and so again i mean i went into research because i was incompetent in anything else you know outside trading so i can play tennis so the first thing i realized looking at a coffee cup is that we had a definition of fragility i was an option trader trading volatility and i realized there were packages that like volatility and packages that hate volatility the same thing was right there staring at me look i mean the coffee cup i looked i was looking at i'm lying was not as attractive as this one it was an ugly coffee cup not starbucks but an ugly coffee cup anyway so i realized that it did not like volatility so here we have the definition of fragility and from there we can figure out anti-fragility if you're fragile to some classes of events you can have the reverse reaction to you know uh to them you can you can gain from from from from shocks so so my definition of fragility came with this i don't know is this this is fragile so with the following very simple thing everything fragile on this planet must have acceleration of harm and this i figured it out from trading like if the market goes down one percent so i'm removing this so people don't read the slide so the if the market is down one percent say you lose a million down two percent you lose another five million so there is acceleration and that acceleration is quite universal and that applies to us if we jump 10 meters we die 10 meters a lot worse than 10 times 1 meter and why is this necessary because if you're linear to harm then you're too harm okay if you look just walking from here to the whatever would kill you because a lot of shocks so we're sort of insensitive to small shocks then we get disproportionately more more sensitive to others because everything that's linear to harm is already gone so everything fragile is gonna be non-linear to harm and we're gonna see everything antifragile would be nonlinear to harm and everything i'm going to be talking about for the rest of my life is based on that very simple point so this is the coffee cup it is not related to harm because i tap on a coffee cup it doesn't break doesn't break doesn't break then a little more it breaks and we're going to see why averages don't matter anymore and and [Music] many things that come with the idea so second order effects is effectively what we got to concentrate on and you can see that really with the pandemic okay the supply chain second order effect is when i put some equations here just to wake people up and let me remove and put pictures so second order effect you can see it easily don't cross the rivers on average four feet deep and translate it into french it's a convex function of an average or whatever so you could also translate into russian or it's even more theoretical so that's what i call convexity effect but you can see it was a supply chain story if you consume zero for one year and 200 units the second year it's going to be a lot worse than if you consume 100 the first year and 100 second year so we have expo we can figure it out if the the the market causes fragility or not in some domains okay unevenness you want steadiness we are fragile to temperature for example i'd rather have 70 degrees today and 70 degrees tomorrow rather than zero degrees today we're not close to it but it feels like it right sort of like today in boston all right anyway zero degrees today and 140 tomorrow you see the idea so whatever is fragile depends heavily on second order effect and that's quite universal so i'm taking this of course you know in 40 minutes i'm not going to be able to cover a lot but i'm going to think this is an interesting territory is that size causes fragilities and it was a very simple example a large stone if someone threw this at my head i i'm you know you'd have to someone else to play you need someone else to replace me for the continuation of the session because i won't be here but if you took the big thing the big uh rock and broke it into pebbles and threw them progressively in my head nothing would happen so you see there is a size effect and pretty much everything in finance everything that breaks bankruptcies all of these come from acceleration of harm because if you're very sensitive to harm in the beginning it would show so it has to be hidden harm has to be hidden and then accelerate just like if you smash a car against the wall a thousand times at a tenth of a kilometer per hour nothing's gonna happen but once at a thousand kilometers per hour you know what's going to happen right so so this is that acceleration and you can see very easily this can explain why an elephant is very fragile if if you push an elephant i mean it's we're not i'm not inviting you it's a thought experiment to push an elephant but if you push an elephant off this stage would break a leg now again i'm not inviting you to push elephants off the stage it's a solid experiment but and when elephants break a leg they're gone this is why we don't have a lot of elephants they're vastly more efficient than a mouse metabolically heart rate 50 versus 200 but guess what a mouse you can you can throw a muscle it won't even know this it will laugh at you all right so uh if you throw a mouse off the stage or something many times it's high at its uh height so there is an effect this economy is a scale that comes with fragility because of that non-linearity and that's also universal we have i don't know about boston but in new york we have at least eight million mice nobody counted them so you can imagine because they don't vote that's not precise but the but you can imagine and you don't have a lot of elephants all right so that explains why small companies tend to survive shocks a lot better than large companies particularly when they are squeezes so all of this now hopefully i'll confuse you a little bit the next two three slides is linked to options because options have non-linear payoff and whatever has this optionality likes volatility and likes uncertainty for some weird reason and then we look at fragility as an inverse of that the other side of the coin is the the the fragile and basically i'm gonna confuse you one more time and then becomes easy all right if you like one of these like them all so if you like and let me explain how if you like uncertainty you will like time why time brings uncertainty you see this time if the coffee cup there's no coffee cup here but if we had a coffee cup on the table you know that overtime is gonna break because you're gonna have an earthquake or something okay so so time uncertainty shocks stressors entropy all of these go in the same direction if you like one you like them all that's why i call them the soda brothers like moving like the brothers that move you know furniture in new york city is the name so sort of rather so the the and and and i spent it took me 10 years to really formalize this uh this idea that if you like one you like them all and and you realize uh and you multiply the examples but what is convex is what likes variability a very simple example of what is convex is what you'd rather have for a lung people don't know that for long ventilators particularly during covid it's something that people didn't pick up on immediately if you give the patient a hundred percent of the dose all the time the patient dies if you give the patient 80 percent and 120 of the needed dose the average dose guess what they do a lot better booster survival so you are convex and concave means the average okay is better than the outcome and basically that's pretty much what it is all right so convex sorry no i can't go back concave all right that's the best way to know it convex means you have an acceleration in your favor okay concave and acceleration against you that's sort of like the general idea and we have illustration and basically if you're concave to randomness to harm to anything you won't do very well in the long run another example this here sitting on a table if you put it on the table for a year we'll experience about what half a million to a million small earthquakes no effect one large earthquake and it will break so this is if you have accelerated harm basically concave is accelerated harm so uh and here our show it shows in medicine where you can have everything in nature has an s-curve where you start by being convex and then end up being concave like for example your frequency of meals it's a lot better to have you develop diabetes from two frequent meals to have say one meal a day than 18 meals a day for the same average but it's a lot better to have one meal a day than for 18 days okay rather than zero meals and then won 18 meals in one day you see so the thing is there's an s curve where we are convex in the beginning and then concave later when you're convex you want unevenness for that segment and when you're concave you want steadiness so coffee cup wants steadiness some things like unevenness we're gonna see businesses good businesses like unevenness chaos turmoil and they're all convex so to repeat why is this connected to an option because again i was a trader an option you have it's convex you gain more from the upside than you lose on the downside say if the market goes up one percent i make x if it goes down one percent i lose maybe half an x an entrepreneur you come up with an idea we're going to see how you tinker okay you make a million if you're right and you lose ten thousand if you're wrong less convex if you make more if you're right for an equivalent thing you are convex okay so this is up to his own okay and and effectively there's something i proved that everything that has a floor would be convex of the floor and everything that has a ceiling will be concave around the ceiling and that applies to nutrition to anything okay and so so a lot of people are working on this in oncology i mean a lot of people a few people i mean i'm saying a few people are working on this and oncology those response because everything follows as i said in nature in s-curve in biology and medicine they model it under some form or another now i spoke about fragility and effectively there's no word to express the opposite of fragility because if you say ask someone there's actually a package other door was fragile written on it but if you ask someone the opposite of fragile they tell you robust or resilient no it doesn't get better from sharks they're robust the robots just like wants you know well it's like people in brooklyn they don't care right you hit it on them they're tough right they they're like don't get better they don't get worse all right so they're not so there's got to be that property that's the exactly mirror image of the fragile that has to have some accelerated and this is hidden in many things and i just try to express it to make it clear in the book anti-fragile and it took about a long time for people to understand that it was not the same as robust and so let's talk about anti-fragility well i i i'm sure that you're familiar with this concept of a gym no now why do we go to a gym because it looks like our body needs stressors you know why why do you have a sauna well because you need thermal some thermal stressor all right moderate stressor and effectively to figure out why there is convexity and exercise okay uh running if you run a hundred meters in say uh 12 seconds you do better than spending then then if you do it over an hour you see or another example if you go to the gym what's better to lift a thousand times something of a few ounces or lift once something of a thousand ounce what's better to lift 100 times something of one kilo or one time something of a hundred kilo depending on we are on a curve in fitness visibly there is a point where it's better and there's an s curve you are convex in the sense that you gain more from lifting five times or ten times something heavy rather than thousand times something light so there is an acceleration so when i wrote antifragile heart rate variability was still not popular and actually i wasn't aware of it but you would ask a cardiologist at a time you know is it good for the heart to have steady uh you know to have a steady in frequency between intervals and they say yes yes it's very important to have a steady pulse what it turned out the best predictive of death is guess what a steady pulse and and and it turned out and now they have heart very variability i have it on my apple watch all right when i wrote antifragile 15 years ago published 10 years ago my apple watch has heart rate variability it wants me to have non-steady heart okay well up to a point so this property but guess what what is the best predictor of a company's bankruptcy the single best predictor for companies have been around safe for which you have time series of five or ten years low variance of earnings and actually even work low variance of cash flows because earnings can be manipulated that low variance of cash flow is a predictor of a death of a company and we actually have plenty of data for funds hedge funds that have no volatility usually go bust in 2009 that was the best predictor of the bankruptcy of a phone so and we're going to see why having some volatility in the system is important and the thing comes from a property that we human have and that and that's a question i put as a title of a chapter on the main difference between a cat and a washing machine on the main difference between something organic or something alive and something engineered the the the the problem i had already with my my uh with with my books is that people most people the black swan had sold millions of copies and was commented on by people who just looked at the chapter headings and thought they could understand it from a chapter heading so i made sure none of my chapter headings could explain anything in anti-fragile right so and the central one is on a main difference between the cat and the washing machine and what is it is that everything organic requires stressors if you're anti-fragile it's not trivial you require sweaters to function properly without spreaders without stressors you you wane but you know what would happen to you if you don't subject your bones to stressors say say someone uh from boston invited you to go on a space shuttle all right and spend some time so there's still some russians up there you can engage in politics okay so say you spent six months up there what would happen to your bone density when it came back to earth when it come back to earth okay so visibly we need everything organic communicates with the environment via stressors so i have a mediterranean complexion so i gained protective uh you know uh tanning i get that i was hiking for four days so i got i tanned why because my body predicts is protecting me predicts more more so i communicate communicate with the environment not because my brain decides hey let's stand because to be protected no because the stressor causes a reaction i get stronger from a stressor bearing weights okay my bones get stronger from the stressors of pounding okay or gravity so the thing about the so the cat if you don't abuse your cat please but if the cat if you put the cat in a you know and then make it idle you know it'll get weaker but the washing machine doesn't have that problem so as a matter of fact if you go bang with your wrist on a washing machine your bone will get stronger okay don't go too hard because you can break something and blame me but but try and the washing machine will not get better so that's one main property of systems so so when i left trading with traders speaking opaque language nobody understands even traders will understand traders so there's no uh unique language so when left trading i realized there's no clear language to describe risk so most people think and and in financial theories a lot of people in boston here you know harvard business schools yeah okay have bad ideas where for them risk is low variability and as you can see from the heart from a lot of things have nothing to do with low variability you have to make a difference the central mist between risk and variability risk is in the tail for large events bankruptcy is a risk but variability doesn't predict that risk and very often for a human body for a heart for a lot of things the opposite so that's a big mistake and effectively i mean i was i was subjected to the war on lebanon and i came out of it everybody asked me about my trauma what do you mean trauma i was a kid during the war and it was it was terrible then trauma no but it turned out it's not universal for people to get trauma for war but there's something the opposite that i discovered that of course people get trauma that post traumatic disorder but there's something vastly more frequent called ten times more frequent called post traumatic growth if the trauma is not too strong thing grows from it just like anything right like going to the gym and guess what nobody talks about it why can someone tell me why because no clinical psychologist is ever going to make any money try to cure you of post-traumatic growth right now you know you're not going to have sessions to mitigate your post-traumatic gains right so this is why it's not it's not studied they're not going to make any money from it sending you drugs for that all right so so i mean it doesn't mean that you have to subject your your children to harm but the moderate amount of stressors is needed so now let's continue when things get stressors there's a mechanism of upregulation upregulation like you subject something to a stressor it upregulates like your attention span if the problem is a little more difficult becomes a lot better like great horses they lose races with bad horses but they win against exxon horses so they upregulate it's well known so there's of course the saying when life gives you a left lemon you have two options in life you have the defensive option there's going to be storms so you go in the basement and cry maybe apply for some government grants okay and or go on twitter and complain whatever it is so you that's one option one option two you say oh is it gonna be a problem let's make some money out of it so that's a pla plan the second option well it turned out that it's a little beyond what i thought now what is the best way for a country particularly if it's maritime to get rich okay let me reverse it right what what is the worst possible way to enrich your country give it resources the most successful countries in history typically are small city-states singapore today is probably the most successful country in modern history i mean except if you like to chew gum i mean you would share my opinion that it's a very well run place they have they don't even have water they have nothing they was thrown out of the malaysia you see the only country known to have been fired okay usually it's like a reverse putin you guys go away all right they had no resources nothing so they were forced to transform but they had water around them water into gold so you give the country gold they'll get poor exactly what happened to the spain yards in latin america you give someone no resources they upregulate and effectively i mean i i know well some some phoenician history over compensation and how the the the the city states venice they were all they had no resources so they had to go get resources so what do you do what did the phoenicians do they had nothing they had the water well they had to go to cyprus because it was very fashionable then to use copper you needed copper cyprus named after copper is where copper was how do you go to cyprus what you need a boat okay how how do you you know make a boat well you cut some wood so they figured out how to make boats and how to cut wood to make palazzos because they built the temple and then they figured out how to trade because if you go to cyprus okay and then you come back you tell them you want some french fries with it or something so you know you come back with so so people started commercial networks simply because they had no resources that's up regulation that come from stressors and this is very i tell you this is very common from companies we see it in the data that unless companies have difficult periods they don't grow so there are a lot of ways you know the chaos monkey also to improve your system is by if you if the world doesn't stress your company out you can do your own stressor you know you can attack hack your own system for example and then of course you can see what happened before and after the dot-com crisis and that's what i'm told i'm told that before 2000 it was this and after 2000 i stopped getting email all right before to you know so my nightmare is emails because now social media it's on social media which is another nightmare but not as as bad and then finally let me explain i said that if you like variability you are convex so evolution works by convexity it's come by is anti-fragile so you're anti-fragile up to a point and let me explain how let's talk about that form of evolution that takes place outside bedrooms you know where it's called parthenogenesis where virgin reproduction it does exist in nature where an organism reproduces itself okay so you're gonna have identical copies of yourself would you no once in a while it's like in the old days when you send a fax machine with instructions and be a little blurred so you have a an error in in printing i'm sending to a 3d printer something with an error most of the time the error is bad once in a while the error is good so you benefit from errors and that's basically how evolution happens it's errors but guess what you want to make some errors not too many why because if you have too many errors you undo the gains from previous errors so there's a sweet spot where your optimal rate of errors and that's pretty much what we have mutations this the other thing happened with a regular reproduction where you have a lot of offspring they're not the same because the random recombination of uh come from from parents and the ones that are better for the environment so dispersion and variability works for you here as well so error and variability are the same thing and that's the anti-fragile so the anti-fat is not just something that bang over on on her or his head the the head gets stronger it's something that benefits from that the sort of cluster and it's the same way it's the same representation mathematically so so now let me explain how we we probe uncertainty okay do you recognize this object so now this is now that i'm in boston and i'm going to talk about what i used to call the soviet harvard delusion all right i'm sorry the soviet harvard delusions and let me talk about it now this is commonly known as a wheel it was discovered supposedly in mesopotamia six thousand years ago why did it take six thousand years for this application those who are old enough here i'm looking i'm finding some gray and white hair will know what i mean [Music] traveling with a suitcase think of traveling with a suitcase without wheels we did that for 6 000 years including a large chunk of my adult life particularly when i was a student was with two or three suitcases right so why did it take so long because we're we're not good at figuring out the future not even what we have in hands like my friend mike showed me something that we discovered accidentally only because we forced this okay 3d printed the thing that unpleasant stuff they give you before you know sometimes going to work is that very unpleasant okay now it's in one piece well they were forced to 3d print them because there was a shortage and and and they discovered the idea that you can do it in one piece necessity is a mother of invention and sort of it's not it's not i mean they never thought about it making them all in one piece before anyway so you can't probe uncertainty very easily we're not good our imagination is not good we're not good the world is way too complex for us now the story of the wheel is actually more vicious than what i just described you recognize these objects from your last trip to cancun i guess no or unless you don't do culture when you go to cancun these are the pyramids they built in the okay uh in the miso americas at a time all right so they built these without the wheel you know it's very simple they had a lot of corn maze corn maze slaves you get this all right you don't need the wheel is it true they didn't have the wheel mexico city museum national museum they had the wheels but they used them only for children's toys they didn't they didn't even make the leap of using the wheels so you realize that we're not good at predicting the future so we're better at doing and the idea is that being convex to uncertainty means having more to gain from tinkering from these errors that make you do well these random errors only because you're convex you have more to gain than to lose from them and visibly there are some industries that are error-loving pharma as you can see i'm i'm i'm sure nobody here has heard but there's a drug so called viagra nobody has heard of it here but you can you can google later see what it does but it was not meant to be something to improve the lives of uh of retirees and you know play golf right that was not the aim of viagra the aim was something else but they discovered that well in the experiment that the the the the the the part of the cohort okay was they didn't have the placebo appeared extremely happy okay so so you make a lot of body so pharma has worked with mistakes and kept going with mistakes thinking that that they went by so the whole idea of design versus tinkering so of course we need both science and tinkering but my idea is that part of a lot of what we do is lecturing birds have to fly in science what we do with our hands is so convex that the example i use is that i'd rather be anti-fragile than smart and i think if your brain assuming iq worked and doesn't assuming your your head has a 120 iq points your hands have a thousand a thousand okay by tinkering trial and error and doing these things so let me take stock and summarize and connect things i've defined fragility i define anti-fragility remember anti-fragility is not uh for for everything like for example if you bang me on a head or if i jump 10 centimeters my ball gets stronger but if i jump 10 meters the only bones that get stronger is those of the people carrying my coffin all right so that was you so you realize it's up to a point you have starvation up to x hours is okay but if you start for two years you know not excellent i mean i don't know you can try but but but i predict it's not going to be very good okay so at up to a point it's better than natural things and now hopefully during the discussion we're going to talk about how i i discovered yesterday that that 3d printing removes some fragilities how does it remove some fragilities well anything optimized is fragile okay so to counter that typically you need redundancies you need to have 18 different uh suppliers or whatever certain number of suppliers make sure they're not the same time zone they don't speak the same language don't have the same culture so they won't have revolution at the same time all right but if you're if you have a backup option to circumvent that with an insurance to be able to print your way out of a shortage then you're not as fragile it's not it's not much more difficult than that okay and i didn't think about it before i had no idea about 3d printing that you can print this beautiful object you can sell it to me and tell me it was made by a very very very fancy artist and i would have bought the story anyway so thank you for listening to me and i'm honored to be here okay talk all right so um i'm going to channel everyone's questions back to the scene here and we can have a little discussion about this i'm glad you summarized at the end because i was taking away from this but i'm supposed to be a mouse that goes to the gym and seek out trauma yes seeking out trauma i love the the ps you know post-traumatic growth uh you like the other syndrome too i'm gonna use that one so i can go to ukraine don't go to ukraine though so um you know i love that also you brought this back to 3d printing at the end and you know more more lessons here for this um group you know we do have a mixture in the crowd of startups that perhaps you know are more nimble or designed to be more nimble but we also have the the bigger companies in the audience that um maybe we could talk a little bit more about the fat tail okay so syndrome you talked about the beginning and how that how that kind of what the lessons are for the bigger companies and fragility and and what accrues to concentration around the fat tails okay so so let me explain the tail story which is pretty much the theme of the black swan that there are two kind of environments it's not let's not try to figure out events let's try to figure out the type of environment we're in let's say that i gather a thousand people randomly and weigh them and i added to that sample the heaviest person i can find on a planet all right so you call up hospitals and see you know all right the heaviest person you can find how much of the total that person represents barely any okay so our weight so what happened you can have outliers but it's not going to make a big consequence well let's talk about food okay tonight you're in boston you're gonna go overeat eat all the mobs or whatever how many cows can consume in a day three thousand four thousand it's nothing because a year is what six hundred thousand to a million depends if you're a new yorker or not depends of whatever you get it so in new york are a million calories a bostonian eight hundred thousand so okay so nothing so a day not one single day is going to make a difference so not one observation would make a big difference but now they are environment where that rule doesn't hold the first one is everything you study in business school the models standard deviation all that is made for that environment i call it mediocre stan the bell curve the gauss curve this that econometrics regression correlations modulation all that that domain the second domain i call extrema stand the same thought experiment a thousand people and you add to that sample the wealthiest person you you can find on the planet as as of today i mean the market has opened yes so we can because it changes all right so the wealthiest person how much again randomly selected thousand people from the planet how much of the total wealth would that person represent not weight wealth so what we have 265 billion versus what an average of 200 000 maybe or not even maybe depends if you have too many people from whatever so so we have a concentration so the domain that delivers this some domain are concentrated that person is wealthier than the bottom 2 billion right so the same so book sales have that effect uh what else are they you know you have millions of novels maybe 10 novels so more than the remaining 999 whatever okay stock market the stock 100 000 companies join the market most of the return over a century long run come from 0.2 but look at the s p 500 sometimes 100 stocks represent half the world the u.s capitalization sometimes sometimes 300 stocks sometimes 30 in in the 80s sometimes 30 companies we have the capitalization of the countries all these technologies worse is the most fat tail 25 000 companies were selling software in 90s uh one of the 25 000 had more than half the sales just to give you an idea of of the kind of the environment so that environment has its rules and you want to be protected from these big deviations pandemics by the way and wars are the fattest tales and and i did right before covid the paper in nature physics on the properties of camden pandemics so the fattest tale pandemics because if you look at how many diseases we've had in history some of them have killed a large proportion of the population so going back to so if you have these fat tails are there lessons for some of the bigger companies in the audience of how to take advantage or how to respond or to build yes so when you're in an environment so it's simple if you go to work as a butcher okay or if you go to work as an artisan you have this predictable income you have this but if you are in a world that is has these properties like for example if you are in technology all right google has what proportion of the internet traffic so it's going to be hit or miss for you and it's going to be either going to be monstrously rich okay or living with in the basement of your grandmother something like that i mean the kind of payoff is it's going to be very extreme in that domain same for companies for sales for everything you have to know about it maybe you want to play the game okay you want to play the game so the the an anti-fragile i talked about the barbell to be robust you spend if you're an extremist and make sure that you have a something that pays your bills makes you alive so the the the strategy typically i i have in a stock market i just have 10 percent of my investments are monstrously speculative and 90 as little risk as i can for an average of medium risk it's a lot better than everything being medium risk it's safer so for example that that idea of having a dual strategy of extreme risk on one side and backed up by extreme safety on the other not the middle for example it may be optimal so i said answer everyone's questions around how to be anti-fragile it's a big company um one question we talked about last night too was the ability for systems to learn or people to learn organisms to learn or not or maybe not having the ability to learn it is from different shocks in the system can you talk a little bit about that because you know there is sort of the psg the post-traumatic okay so for example growth opportunity so but then how do you learn and continue that okay so this is uh going to be not good news on on many in many respects for the following take the the restaurant business the restaurant business is the most uh uh what was very telco the most viable business we did not have a bailout to restaurants in america right it is so but the problem is the restaurant business doesn't learn because chefs buy a book and you know and and and learn from that book it learns via bankruptcy that's the the the idea that systems learn by filtering by evolution and that is so the system the restaurant some food in new york is vastly better than food in soviet russia okay i don't know if you've been to soviet russia okay or post-soviet russia you can compare food why because you have the system in new york learns to make good food via competitions what we call scale transformation so you need the the bankruptcy of the restaurant to improve that system likewise in technology you need the bankruptcy of the company for the system to work but the thing is for you guys involved in technology individually the idea is to fail early you don't want to fail late and and it's healthier to fail early and this is why we never had a bailout of technology in america we only have we have bailouts of banks of uh of of of companies that don't have this structure so as we think about second order effects you know we talked about the supply chain and if you are small organizations or if you can be nimble and anti-fragile if you can be that mouse you can react more quickly but there it's all tied together can you put this in the context of larger economies and specifically related to manufacturing and supply chain how you can sort of anticipate events and how they cascade or if you can okay so so the the let let me uh get to it via globalization right so people don't quite realize that the term globalization is meaningless it's ill-defined okay they are definitely comparative gains from trade for some countries to produce wine and other to produce apple juice okay so and it makes no sense to have salmon in saudi arabia to produce salmon saudi arabia or to produce camels in norway all right so it makes no sense so the the plus there's this cross breeding of ideas that happens and often people you have in new hampshire uh criticizing realization using what twitter so but so it is really defined for the following reason some you know on one side you have otarchy which means that each one lives off of her his land and garden that's it and closed systems and no trade okay on the other hand you can have completely open borders which actually now start to have thanks to zoom because there's no visas i don't know if they they've started to have visas to be able to you know get on a conference call on zoom but so far the the the state department doesn't know about zoom very well i mean doesn't know about it so you don't need visas to get on zoom so we may have open borders uh in cyber space anyway but physically we don't have so when people are pro or against globalization it's a matter of degree someone a little more than others it's a matter of degree somewhere and there okay so but some of the ills of globalization poorly done is when you buy everything from one supplier okay everybody buys everything from one supplier before the recent wave of globalization that started in the 90s okay you used to have people making something the product in arkansas you can even probably buy it in in florida and and argentina and stuff and then progressively that unit will be made you know where all right in one and even in one factory of that place okay so start depending more and more on that particular country and within that country on a particular district so if the boats cannot get to that country you know it can happen as if you see the news today there's some country that has a lot of or actually you showed it the boats you know that that can't load merchandise so the whole thing stops okay so that is one fragility that comes with globalization that remember it's the same process that got a billion people or two billion people depending on who you listen to out of poverty so it's not like it's a bad process you just have to make sure that you reduce your fragility okay find your facilities and reduce them that's what what what the situation is globalization is i mean been going on since ancient times this is why we had the silk road it bankrupted rome okay but i don't know if it's good news or bad news but bankrupted rome because they got older uh they got silk and then and they you know they bankrupted them but but at the same time they have this i mean i'm sure that if you go to mit you walk around you will see a lot of variations and first names no it's not gonna old you see you had a lot of countries represented this is globalization okay we're basically cross-fertilization of ideas so it is so let's not uh but people are imprecise when they talk about globalization i was actually sitting on a panel like this and what i figured out that people had different definition of globalization the russian there was a russian minister of trade or commerce something like that who spoke he said i'm going to speak in russian because of globalization for him globalization means everybody should speak russian and then there was a french prime minister fiona who was then ex-prime minister who said i'm gonna it's globalization therefore i'm gonna speak in english i'm a frenchman i'm not speaking so you can interpret it the way you want the idea is let's figure out what it means and the side effect namely supply chain shortages which apparently if if they can make this beautiful black swan or they make so many different objects we have an insurance policy visibly against shortages maybe maybe progressively plus it's good to have shortages why is it good to have shortages because it forces us to substitute to find substitution to do things to to think out of the box just like the you know the errors that gain make the system gain from it that's great all right well thank you for you for sharing your philosophy behind your numerous books and i'm sure many people have questions here but we can find you during the break and and cover you with those we have a short break coming up now and will reconvene at 11 is that correct okay 11 o'clock back here with carl boss great thank you very much extremely honored to be here

Channel: Formlabs

Views: 66,319

Keywords: the digital factory, digital factory, nassim nicholas taleb, nassim taleb, nassim nicholas, antifragile, antifragile nassim taleb, antifragile nassim taleb summary, antifragile nassim, anti fragile, antifragile things that gain from disorder, antifragile book, nassim nicholas taleb book, nassim taleb book, book author, book, author, ted talk, ted talks

Id: g2LYTUEOn20

Length: 55min 15sec (3315 seconds)

Published: Tue May 31 2022