Q&A – The Future of Humanity – with Yuval Noah Harari


I’d like if I may to to start with one myself and it picks up on some something that I thought was very interesting at the end of your book when you’re talking about some of these issues and you say one of the issues that we have to think about is that with these advances in AI intelligence will have become and that would perhaps already has become decoupled from consciousness you haven’t talked about consciousness and that’s one thing that I guess at the moment at least we feel we have that machines and AI doesn’t have of course there’s a debate about whether that will ever change but at the moment it is something that distinguishes us but I suppose what you said has raised the question is there actually an economic value in consciousness what where do you think consciousness is going to feature in these questions of who will do what well I think consciousness at present is the biggest riddle we still haven’t solved when it comes to understanding life or to understanding the universe a lot of people especially in the AI business tend to confuse intelligence with consciousness but there are very different things intelligence is basically the ability to solve problems and a consciousness is the ability to feel things to have emotions and sensations and subjective experiences now in humans and in all mammals intelligence and consciousness go together the way in which mammals including humans solve many problems is through feelings through emotions emotions are not the opposite of rationality of intelligence they are the embodiment of evolutionary rationality so in humans they go together to such an extent that many people think they are the same thing ah but it’s not the same thing we have other organisms that have sometimes high intelligence without any consciousness these are for example many plants many plants trees and and so for they are highly intelligent in the way that they solve certain kind of survival problems but as far as we know they have no consciousness they have no feelings they have no emotions they don’t feel pain you don’t feel love they don’t feel fear Computers an AI in this sense are much more like trees than they are like baboons or giraffes or Homo sapiens they are running not only just toward intelligence but about super intelligence while bypassing the paths of consciousness they are taking a different route to intelligence than the route that mammals take and over the last say 60 70 80 years there has been a tremendous progress in computer intelligence there has been exactly zero progress in computer consciousness in the days of Alan Turing a computer’s head zero intelligence as a zero consciousness and today the best computers in the world still have zero consciousness as far as we can tell they have no feelings they have no emotions and despite what you see on in science-fiction movies there is no indication that they are going to develop any kind of consciousness are anytime soon there is no reason to think that they must develop consciousness if they want to become super intelligent in most science fiction movies there is this fallacy that are the matrix or whatever artificial intelligence is fighting against humans if it’s going to be really intelligent it also must have consciousness and this fallacy appeals to us because we are mammals and in mammals intelligence and consciousness go hand in hand but computers are not mammals they’re not animals there are something completely different so what we may see in the 21st century is the development of a super intelligent but completely non conscious entities and this is quite frightening if I spoke in the beginning of the talk about the potential of artificial intelligence of even expanding from planet Earth to the rest of the galaxy so we have this potential of creating a galaxy even in you full of intelligence but devoid of consciousness and this is a very scary thought okay I want to open it up to questions and I will try as far as I can to keep an eye on the gallery as well but could I have yes visa throughout history in certainly today a majority of jobs are not fulfilling in any way they’re only important because they give access to the fruits of production they bring you money and so surely the problem is not that humans will not be required to do jobs that human that so machines can do better and quicker and more efficiently and cheaper but that at the moment there’s no means of the distribution of the fruits of production except through work surely that is an important problem not that people will in your words be useless hmm well I used the term useless as a provocation of course and I when I say that these are useless humans it’s not from the viewpoint of the mother of the wife or the thought of a son it’s from the viewpoint of the political and economic system now we could create an unprecedented economic and political model which for the first time in history really decouples work from having the benefits the economic benefits this has never succeeded before in human history which doesn’t mean we shouldn’t try we should definitely explore new models because the old models are fast becoming irrelevant the danger is that if you become economically useless sort of from the viewpoint of the political system you’re also losing your importance your power and this is a very dangerous position to be in if you look at what happened in the night in the eighth in the Tauri in the twentieth century what you see is that governments throughout the world not only liberal democracies but also in dictators it’s invested a lot in the health and education and welfare of of ordinary people even Nazi Germany as long as you weren’t a Jew or something like that invested heavily in your education and health and welfare and built hospitals and schools and gave vaccinations to everybody in so forth why not because Hitler was a very nice person but because Hitler knew in the Nazi elite knew that if they wanted Germany to be a strong nation with a strong army and a strong economy they needed these millions of poor Germans to serve as soldiers in the army and as workers in the factories and the offices now if you take this usefulness out of the equation then the danger is that the state and the elite will stop investing in the health and in the welfare and education of the masses it doesn’t have to be like that but from a historical perspective it is a very big danger now there is talk today about a potential solution which is a universal basic income that say the government will somehow manage to tax Apple and Amazon and Google and all these giants who are extremely creative in in tax evasion so somehow managed to tax them and use the revenues to provide universal basic income to everybody which is a model worth exploring but it’s not an easy model because it’s not clear what universal and what basic mean when it comes to universal so what is universal universal is only citizens of the UK only citizens of the EU are also the citizens of Bangladesh and Somalia and Sudan and so forth we now have a global economy if we find a solution that is good for the citizens of the EU but leaves everybody else outside this is a very problematic solution the second question is what does you need what does basic income means what are the basic needs of humans the basic needs of humans keep changing all the I mean previously people thought oh if you just have access to food and shelter that’s it that’s basic needs but today certainly in the West people think that you need far more than just food and shelter if you have your basic needs met people say even access to the Internet is now a basic human need if in the 21st century you have revolutions in biotechnology that enable you to extend human life far beyond 80 or 90 will this be a universal human need that everybody are entitled to or you’ll have a society in which you in which you have ordinary Homo sapiens with ordinary health and bodies living until 80 and then you have an elite of superhumans with all kinds of super abilities living indefinitely so I don’t know what the answer is but certainly it’s worth exploring are these new models and we don’t have a lot of time to explore them evoke an eye and perhaps expand on that a little bit because it caused me from what you say that and from the question that underpinning some of this are quite are issues of morality of what we feel is a moral solution to this and I can’t help wondering whether some of those you know we have a sense that perhaps there is something intrinsic in our moral instincts but I can’t help feeling that from what you said it’s certainly a lot of aspects of morality perhaps emerge from an economic model of the idea that you know it is that we should work the Protestant work ethic is a you know as moral position that sort of reflects the fact that otherwise you won’t have any economic worth than perhaps it follows from the economic model so where do you think morality is likely if at all to figure in all of this it certainly if it’s the case that morality sometimes emerges from our economic structures rather than informs them how is that going to to figure in the future oh we definitely have to put a greater emphasis on on morality and on ethics the the economic argument will become weaker in the 20th century you had to basic our arguments why governments and states need to liberalize the economy liberalize the political system protect human life human rights protect the rule of law provide health and health care and education and so forth one was the ethical argument this is the good thing to do the other was the neoliberal capitalist argument that if you want to be a successful economy then you need to liberalize the economy you need to release the creativity of people you need to put to have the rule of law to protect property and so forth and as he story and I think it’s it’s a sad truth but it’s still a truth that the economic argument was far far more important than the ethical argument the reason why countries say like China or like Turkey liberalized their economy and their and their society was not that they became convinced of the ethical argument it’s that they became convinced of the economic argument and once you take away the economic argument say in 21st century once you no longer need to protect human rights or to provide universal health care in order to have a thriving economy then I’m afraid that the ethical argument by itself will not be enough right thank you yes other questions please yes it’s one yeah if you wait to the microphone gets here and well here it better you’ve posited an accelerated development of both organic and inorganic life and suggested that in organic life has an advantage certainly outside the planet there’s organic life have any advantages or we’d like you to end up even if we’re super humans as the pets or zoo attractions of inorganic computers well at present we know of one big thing that organic life has and in organic life does not seem to have which is consciousness are the mind and the problem is we don’t understand consciousness in the mind very well we’re making amazing progress in understanding the body and the brain and intelligence we’re making far less progress in understanding the mind and consciousness we don’t know how we assumed biologists assumed that somehow mind emerges from brain that when billions of neurons in the brain are firing in a particular pattern then somehow out of this electrochemical storm subjective experiences of love and hate and pain and joy emerge and we are becoming quite good in our recognizing and correlations that when this type of electrochemical storm is happening in this part of the brain we know that the person is probably feeling angry and when another type of electrochemical storm is happening the person is probably feeling fear but we have no idea how the firing of billions of neurons create a subjective experience and we don’t even know what could possibly be its function why do they know in addition to all the cascades of electrochemical events in the brain why do you need this second layer of mind and mental events we just have no idea and we are not investing enough effort in that direction and the danger is that we’ll end up upgrading bodies and brains and losing our mind in the process without even understanding what we are losing it and I when computers came about the prediction was that we will work three day weeks I’m still waiting for that the reality is that were spending long days behind the computer in jobs so the new jobs which have been created which is not create anything like media jobs I work in media so I’m allowed to say that and don’t you think there’ll be new jobs created and secondly those jobs which are not jobs won’t they be doing something of a higher caliber something that AI won’t be able to do in the end of the day who will create a new IBM Watson hmm well there could be a lot of new jobs humans are very good in creating such such things but the question is whether humans will be better than the computers in the jobs because even their AI is gaining on us you already start seeing like the first articles in sports and in economics being written by algorithms and things that ten years ago our people all laughed at Google Translate but it’s completely useless and it makes make these terrible terrible mistakes so it’s still very far from what a human translator can do but in many cases in many texts especially more technical texts you can put the Chinese text into the Google Translate and you get a first sense of what of what is there and who knows what it will be in 20 or 30 years so the big question is not creating new jobs it’s creating new jobs that humans still do better than AI and frankly we just don’t know where for the first time in history we are in a position that nobody has the faintest idea how the job market would look like in 30 or 40 years if you go back say to the Middle Ages or even to the 19th century so there are many things you didn’t know about the future at any time there could be an invasion from the Vikings of the Mongols might invade you could have a flood you could have a plague lots of unexpected things may happen but the basics of human life including the job market you knew how the job market would look in 30 years which is why you trained your son or your daughter are in the appropriate girls if you’re a peasant you told them how to harvest the corn and how to grind the corn and how to bake bread because you knew these are the kind of skills they will need in 40 years today with absolutely no idea what to teach children because nobody knows what kind of skills they will need in 2050 do you have any suggestions or recommendations in mind I think the best bet under current conditions is to invest in emotional intelligence and emotional resilience because the one thing we are certain about the world of the 21st century is that it will be extremely chaotic with a lot of changes and for the first time in history people will have to reinvent themselves throughout their lives in the previous model of life which held true until the 20th century including the 20th century was of the life divided into two periods in the first period until your teens and 20s you mostly learn and then in the second period you mostly work you make use of what you learned in the first period this is will no longer be relevant ah by the time you’re 40 or 50 most of what you learned in your teens is completely irrelevant so the basic skill you will need is a lot of emotional resilience and the ability to keep learning to keep changing to keep reinventing yourself throughout your life which is very very difficult right I notice there are probably some teenagers here but you may need to stick keep going school moment that’s with a pinch of salt have questions more questions yes are there I’ll take yes yes please yes a little bit and then and then I’ll thank you could we return to the subject of consciousness because you’ve implied that consciousness is a direct result of neuronal activity now there is a whole spectrum of consciousness from basic stimulus and response which could be said to be the result of neuronal direct result of neuronal activity and at the other end of the spectrum is human self conscious awareness which is not the direct result and cannot be reduced to neuronal activity it results from humans living in a community using a language in which to relate to each other formulate and solve problems personally I don’t think at present we have a good scientific theory or explanation for consciousness a lot of people in the life sciences adhere to the dogma which is at present just a dogma that somehow consciousness all consciousness emerges from the Ronal activities in the brain but we are very far from having a theory which explains how it happens and I wouldn’t be surprised if during the 21st century we come to the conclusion that this is not the case that you cannot reduce mind to brain and that you cannot reduce consciousness are to just normal activities in the brain which doesn’t mean that we then go back to traditional religious visions about there is a soul there is a spirit no I think we need to go forward and develop this branch of science much more deeply and carefully are the science of consciousness as against this size the brain sciences and again as I said I’m afraid we are not doing it fast enough when we are investing a lot are in the study of brain partly in the hope of deciphering intelligence and creating super intelligence and we are not investing enough in the study of consciousness and subjective experiences you you made the point earlier and you make it in your book that it’s still not clear what value consciousness has that you know even for social cooperativity there are creatures ants do it or that we toom don’t have a high degree of consciousness which is interesting because I guess often the assumption is that there’s clearly an adaptive value to us being conscious but you feel that we haven’t established what that is or even whether they’re necessarily is one yes again the common assumption there must be some some valid optic value in that and people say oh it’s obvious we need to feel fear so if the lion comes we run away but as we study the brain more and more than we understand what is the biochemical cascade that leads from the vision of the lion from the activity in the eyes how it go through billions of neurons and ends in our legs moving and we’re running away from the lion what we don’t understand is why in all this biochemical cascade we also need the second layer of actually feeling fear why not why can’t it all happen if you take like a billion dominoes and one makes the other fallen and so forth they don’t need consciousness into it in order to do it so why do the billions of neurons in the brain what does consciousness do in addition to what the neurons are doing and we don’t have any explanation for that at present I think the best theory we have is that consciousness is some kind of mental pollution then it’s just an inevitable byproduct of the firing of the neurons which does absolutely nothing it’s like somebody a comparator I think to the roar of the engine of a jet engine when if plane is fire is flying the noise it doesn’t do anything got needle it’s just there so also consciousness it has no practical purposes at all it’s just a kind of pollution created when these billions of neurons are firing I personally am NOT very feel very good about this explanation ah I think we just have to keep on researching could I take the next question going back to the society with a economically useless people as you call them presumably these people will still be consumers so consumerism will have to play part in the way the society is structured I wondered what are your thoughts about that yes one again one theory is that humans the ultimate value of humans will be just as consumers that will do nothing useful at all X but the economy needs consumers however you could have consumers which are not humans which are not conscious in a very simple system you can have for example a corporation managed by algorithms which mines or iron and nickel and and so forth and another corporation which processes metals and produces robots that then it sells back to the mining corporation and the mining corporation sells the the product of the mind the robot corporation and they trade and they make billions of dollars and you don’t need any humans as consumers nobody but this is how the economy function what is the product of the economy to present nothing I mean if you look at the economy from outside it’s just a closed system it does nothing from the outside so if you have an economy it’s of course a very simplified version but to make it easier to think about it if you have just these two corporations trading with one another and mining more and more ore and sending spaceships to other planets to mine I run on Mars and Pluto and so forth you get an expanding and very successful economy without any human consumers without any consciousness which can go on indefinitely and even today we are starting to see this are phenomena of non human or non conscious consumers if you look at the advertising industry so the most important consumer the most important client of the advertisement industry today is no longer human beings there are algorithms and above all one particular algorithm which is the Google rhythm if you want your product to be successful if I want my book to be successful my number one client is the Google algorithm and I know this for a fact because when they asked me to write this blurb or something describe the book in 10 words or 20 words so they have these experts which know what the Google algorithm pays attention to and they say no don’t say it like this say it like that because then the Google algorithm will will notice you more so we are already in a stage when at least in some areas the most important consumer is a non conscious algorithm that’s very interesting there’s a question right at the background so I can’t see if even if you’re male or female but yes this now I can and don’t you think that human beings in the future won’t be dissociated from AI but instead will be kind of augmented human beings using AI to be more performance during in our jobs and instead of working for 8 hours will be then working only for 3 hours which will leave us more time which will free us further to do whatever we want in life like singing yoga you know arts painting whatever yes that’s definitely a possibility that will create this not useless class but layered class and people will have far more free time and so forth again part of the problem is that if you look if the state gives you the necessary conditions the food the shelter everything then you can spend all your days just listening to music or creating music or whatever but it’s for a historical perspective it’s very dangerous to be in a situation when you are expendable when the system doesn’t need you now we could see a merger of humans with AI of humans with computers and it’s clear what humans will get out of it it’s less clear what a highly intelligent eh I might get out of such such a merger so I’m not saying that it won’t happen it’s another possibility worth exploring but I all the time try to get back to the basic social and economical and political reality because I think and you mentioned it at the beginning of when when you introduce this subject that a lot of futurology at present is far too much fascinated by the technical aspects of AI and robots and biotechnology and we can do this and we can do that and it’s too divorced from the are gray day to day realities of economics and politics and technology is never deterministic the mere fact that something can be done through a certain technology does not guarantee that it will actually be done it will actually be happy which will actually happen with the same technology you could create completely different societies if we look back at the Industrial Revolution so is the same technology of trains and electricity and radio you could create a communist dictatorship you could create a fascist regime and you could create a liberal democracy the trains and the electricity didn’t tell you what to do with them so similarly in the 21st century there are all kinds of possibilities that are opened up by biotechnology and AI and so forth but it’s not a terminus t’k and at the end of the day they’ll still being political and economic decisions that will determine will decide what the direction will be sorry thank you there’s a recurring theme of rate or speed of change and that’s bringing about an idea of urgency can you just talk about whether humanity is able to deal with this rate of change when we think about historical experience historical evidence one of the big problems of what’s happening now is that the rate of change is accelerating and I think that and for the first time in history technology is outpacing politics I mean previously even in the 19th and 20th century despite the amazing technological changes politics was still able to keep up to understand what is happening to understand what are the potential implications what are the various models various solutions we can adopt and you could still have a political process deciding what kind of society we want to create using the new technology now technology seems to be outpacing politics so the political institutions and the political processes are left far behind and they have less and less influence on the direction that the technological revolution is taking to give a concrete example over the last 20 years maybe the most important change in the world was the rise of the Internet this changed the daily life the economy the society the politics more than almost anything else it touched upon traditional political issues like state sovereignty like security like privacy but the decisions about the shape of the internet will not taken by any traditional political process there could have been different kinds of Internet’s we are familiar with the version that emerged but this is not the only possible way to create an Internet decisions were made in the 1980s in 1990 90s about the shape of the Internet but not by the traditional political process I never voted about the shape of the Internet it was never an issue in any political campaign in any elections in Israel and I think this is true of all countries are very few political parties have any policy on on the internet how would it would look like maybe they have policies on very particular points like who controls the data or taxation of the tech giants but they don’t really have any serious vision about what they want the internet to look like and this is will only accelerate in there in the coming decades so already today were in a very curious and frightening position when we don’t have any political vision for the future traditionally politics was a battleground between visions different visions about the human future of course there was a day to day issues of all kinds of sex scandals and things like that but the big thing about politics is that you have two or more groups with very different visions about the future of the country or the future of humanity and the political process is meant to find some kind of pop off of a solution or compromise this is how it was in the 20th century when you had these huge visions the communist vision the fascist vision the liberal vision everybody had this big vision about the future of humanity now nobody has any serious vision about the future of humanity except perhaps for a few silicon gurus in in California uh neither the right nor the left has any meaningful vision of how to use biotechnology in artificial intelligence in order to create this kind of society or that kind of society if you look at the elections in the United States at present so you don’t hear anybody talking about it Donald Trump has no views as far as I know about artificial intelligence and it’s likely not because he’s hiding something I don’t think he’s capable of hiding I don’t think he’s capable of hiding some very important views keeping me to himself it’s just not part of the political political landscape which means either of two things that nobody is thinking about it or that it has become the monopoly of a very small group of Engineers and enterpreneurs in places like Silicon Valley and it’s good that they are thinking about it at least somebody does it but it’s bad that nobody else is doing it especially because they don’t really represent anybody except themselves so I think what is very urgent is to start a political debate about these issues before it is too late and politicians and also the public need to need to move faster before technology completely outpaces them and they lose their ability to do anything about the future of humanity in the future of life do you think it’s also the case that political philosophies in general have been based on particular ideas about human nature those ideas might differ but the political philosophy starts with an idea of what humans are like but it seems that we were more mutable than that and we are changing our behavior in response to technology so perhaps is that one of the challenges – and they relate to this question of change and how we respond to it is that one of the challenges for formulating a sort of political vision but actually you know because of the way technologies have gone we are doing things we are doing things now that we wouldn’t have imagined we would be doing 20 or 30 years ago however factoring the mutability of or how mutable is human nature I suppose I think this is really a key issue maybe the key issue because previously in history you had like two sides of the equation you had things changing are in technology in economics in society but then you had an immutable human nature the assumption is that this does not change so the political question was how do you given a constant human nature how do you make use of a new technology like steam engines or like electricity and so we changed the world outside us and there was a big political debate how to change the world outside us whether to create again a communist Society or a liberal democracy but the basis was that human nature itself is the constant now when we start having the opportunity to change human nature we don’t have any constant on which to stand and to say okay I want to change human nature in such in such a way and in this sense politics is still stuck in the old 19th century 20th century debates and it is not coming to terms with the real revolutions of the 21st century right I have time for one could we take a right at the back there yes edition Honda I think we’ll probably have to call that it yes hi thank you and as for the American presidential debate there actually was one transhumanist candidate Zoltan Istvan who probably most of the people haven’t heard of but he was actually a very technical and to date who was sidelined very early on but more more to the point and can I get your opinion about this idea of liability and kind of dispersion of liability because one of the things that’s stopping those doctors and your phone from being taken up by wider society is the idea of trust and liability so I want to go to a physical doctor because if something goes wrong I can always sue him you know restorative justice and whilst with an app we who do I go to who do I see who doing it my retribution from and we do see this issue of liability with drones you know all those technologies whereby it’s all dispersed so is it the billionaire’s is it to share shareholders who will have to take the hit do we have a massive fund to absorb all those claims people will be just scared of trusting technologies because what if something goes wrong what’s the recourse oh yes that’s a very very good point I think in many cases the major problems are not technical at all the major problems are legal and political this is why it’s so important when we have these discussions about the future about driverless cars and so forth that it’s not just an issue for engineers explaining as technical stuff the really interesting issues relate to legal issues and legal issues of course relate back to politics and to ethics or to morality but our solutions will appear I mean once we start once the potential is there it’s definitely there you’ll start seeing more and more new models of how to solve these are legal issues it will provide at least some work to some humans lawyers and then politicians until AI will take over even that the interesting the really interesting thing is that I think ethics and philosophy will become much more important than in any previous time in history because they will have far more practical implications a lot of ethical debates that have been going on for thousands of years between philosophers had had very little impact on actual reality will suddenly become practical questions of engineering and I’ll give an example mmm with driverless cars so if somebody here ever studied philosophy at college or university so there is all these annoying trolley problems you know that first year class in ethics you have a trolley and it’s going to hit five people but if you sidetrack it it will just kill one person or you can uh drop into a percy piece and you kill yourself and you save the five people so what do you do now philosophy school students and philosophers have been arguing about it for thousands and thousands of years with very little practical implications about how people actually behave because even if you say in theory yes this is the right thing to do when you actually find yourself in that real life situation you very often behave in a completely different way because most of human action is not really determined by intellectual beliefs or even religious beliefs they are determined by completely different mechanisms now it suddenly becomes a much more important issue when you have driverless cars because in a human even if you tell this human being every wicked church for thirty years you must behave like this when real life situation occurs he or she may do something completely different with an AI with an algorithm to drive the car if you program it to behave in a particular way you have a 99 percent guarantee that it will indeed behave in such a way so if Tesla Google or Toyota is now engineering the first fully autonomous drivers car they need to know how to program the algorithm if indeed the car is driving on the highway and suddenly the car notices the algorithm the AI and notices that it is about to run over five pedestrians and there is no way to avert this except by swerving to the side falling off a precipice and killing the single person inside the car that owns the car what how should we design the algorithm this is a practical question of engineering this is not a theoretical question of philosophy there could be all kinds of solutions given that we live in a capitalist neoliberal society I guess somebody up there in Google or Toyota will say let’s just leave it to market forces will we’ll design and build new car insurance will build the Toyota altruist that photo for per stirpes and we also go on the market with the Toyota egg waste that drives over these five people and the customer is always right the customers will will decide so in this way these philosophical problems will become engineering problems well if I bought this you

100 Comments

Add a Comment

Your email address will not be published. Required fields are marked *