(crowd applause) - Ladies and gentlemen, dear friends, on behalf of EPFL it's an immense pleasure to welcome you for this very
special evening on our campus. 2019 is a very special year for EPFL. We celebrate our 50th birthday. And in the world of universities, 50 years old we can think of us as if we are still a little child. And for those of you who have children you know very well that one thing that kids are very relentless
at doing is asking questions. And we find ourselves today
as a Technical University at the very heart of some of
the most pressing questions of the century. There is no denying that
technology has inserted itself into the very fabric of our lives. And so this evening is a
good evening to ask questions and maybe to try to provide answers. So we've assembled a very
particular team of people, thinkers, to try to lead
us towards the path of where technology's guiding us, what is the relationship between human beings and technology. And to explain you a little bit more about how this evening will unfold, please join me in
welcoming Leila Delarive. Leila (crowd applause) is the founder of the
Empowerment Foundation and our partner for this evening. - Good evening, Professor. I think I'd see you again in two hours. Good evening everyone. My name is Leila Delarive. I am the founder and chairwoman of the Empowerment Foundation. We are living at a
tipping point of history. Today we are giving more
and more away our freedom to algorithms, to data. We are living with robots, and maybe one day we will
become robots ourselves: we are talking about augmented humans. So we live in this particular
moment in our history where we can choose
between a utopian scenario or a nightmare. But the choice is ours. The Empowerment Foundation
recognize that public utility is advocating for a human
centered technology, meaning that technology should
serve us and not enslave us. We really believe that
technology will lead to a more inclusive society where each and every one will have more happiness, wealth, and we hope we could live in
a more healthy environment. But still tonight there are
lots of issues to address. And we have partnered with EPFL together to welcome one of the greatest thinkers of this 21st century. He has brought three books
that were sold worldwide. "Sapiens", "Homo Deus" and "21 Lessons for the 21st Century". He is a Professor at the
Hebrew University of Jerusalem and he will help us to understand what the future of humanity
will be, we hope so tonight. Before I call him to join me, I will just give you a few information on the program tonight. We will be joined after
Professor Harari's keynote by different participants. Among them Ken Roth, the
Director of Human Rights Watch. We also have Professor Effy Vayena who is coming from the ETH in Zurich, Professor Jacques Dubochet
that you all know. And we will have two leading
Professor of the EPFL showcasing the research,
Professor Courtine and Professor Bloch and Professor Paik. And we'll be all
discussing tonight among us and you, we hope, to
have this debate about the impact of technology. So please join me in welcoming
Professor Yuval Noah Harari. (crowd applause) Good evening, Professor. Thank you.
- Thank you. Thank you everyone for
coming to this event to discuss our future, hopefully our common future as humans. And maybe the most important thing to know as human beings in the 21st century is that we are now hackable animals. And this is the result of the combination of two enormous scientific
and technological revolutions as a revolution in infotech
and the revolution in biotech which in the past decades
have evolved separately but are now combining
to a single revolution, which, as I said, results
in really the ability to hack human beings. There is a lot of talk
about hacking computers and smart phones and
emails and bank accounts, but the really important ability is to hack the human animal. And this is based on the insight
that is coming not from AI, not from infotech, but
the insight that is coming from the biological sciences that organisms are really algorithms, and therefore algorithms
can hack organisms. For those who like equations
or understanding reality in the shape of equations, then the equation I
can offer to understand what's happening in the world right now is B times C times D equals R which means biological knowledge multiplied by computing power multiplied by data equals the ability to hack humans. Now to hack a human being
means to understand, let's say me better than
I understand myself. To understand what I feel,
what I think, what I want better than I understand it. And once this is possible, it means that whoever understands me better
than I understand myself can not only predict my
decisions and choices but can also manipulate
my decisions and choices and increasingly take
decisions on my behalf. It means the shifting of authority from humans to algorithms. Now a lot of governments and corporations and institutions throughout human history had this ambition to
understand and control humans. But it was never really possible because they never had
the biological knowledge, the computing power and the
data necessary to do it. Even just a few decades
ago let's say the KGB in the Soviet Union, the KGB could follow you around everywhere 24 hours a day, observing, recording who are you talking with,
where you go, what you do. But the KGB did not have
enough biological knowledge, enough computing power, and enough data to really decipher what
was happening within you, what was happening inside your body, inside your brain, inside your mind. Now for the first time in human history, and if not now then in 10 or 20 years, at least some corporations
and some governments will have enough biological knowledge, enough computing power, and enough data to systematically hack millions and even billions of people. And if this happens and if we
don't take countermeasures, this could mean the end
of liberal democracy as we've known it, and the end also to free market
economics as we've known it. Liberal democracy is based on the insight that the voter knows best and that the voter is
the ultimate authority in the political field, and free-market economics
is based on the idea that the customer is always right. That the ultimate authority
in the field of economics is the desires of the customers. So the government, the
political government, should represent the will of the people, and the corporations
should serve the will, the desires of the customers. But what happens if the
government and the cooperation cannot just anticipate the will and desire of the voters and customers but also manipulate and control them? And this is not a hypothetical question, and questions about human agency and about the very meaning of freewill, whether there is such a thing, have bothered philosophers
for thousands of years. There is nothing new about
the philosophical arguments What is new is the technology. We now have or we will soon have the technology that
will enable governments and corporations to manipulate and control the will of the voter and the desire of the customer like never before. And then who represents who? It's not clear. Again, I don't want you to
think about it as a kind of doomsday prophecy because
it's not inevitable. Technology always gives us options, not an infinite amount of options, but are always different options. You can use the same technology to create very different kinds of societies. We saw this in the 20th century when with the same technology
of the Industrial Revolution with trains and electricity and radio and television and cars, some people created
communist dictatorships, other people created fascist regimes, and other people created
liberal democracies. They all used the same technology. It's the same with the new
tools of the 21st century. Information technology and biotechnology can be used to create completely different kinds of societies,
really all the spectrum from paradise to hell. The important thing at
present moment is to realize the true potential of these technologies and to really start the political
debate about these issues. Engineers and scientists
in places like this may realize the true potential of the new inventions
of the new discoveries, but the political system
and most of the public still hasn't realized what we are facing, what the new inventions and
technologies really mean. So it's the job of
historians and philosophers and social scientists to form a bridge between the engineers and the geneticists and the biologists and the general public and to really change
the public conversation, change the political conversation. I think that this should be one of the most important
issues in every elections around the world, in
every public discussion around the world. And what I see as an
historian unfortunately that too much of the political discussion in most countries around the world is focused on the issues of the past and not on the issues of the future, and that too many
politicians are simply unable or unwilling to form a meaningful vision for the future of humankind. If in the 20th century politics was a great battle between visions for the future, good, bad, that's a different question, but in the 20th century
it was very obvious that politics was about the future. And you had the communist
vision, the fascist vision, the liberal vision, and
the political struggle was a struggle between these
visions for the future. Now almost nobody in any part
of the political spectrum offers a really meaningful vision for where humankind will
be in 20 or 30 years. What most of what they
offer is really just nostalgic fantasies about going back to an imaginary past. And this is a very,
very dangerous situation because it really means that maybe the most important decisions
in the history of humankind are taken either by a
small group of specialists who represent nobody or they are not taken by anybody. They just happen. And this again may be part of this process of shifting authority
from humans to algorithms. In 2019 it is still,
humans still have agency but we don't have a lot of time. Within our lifetime this
shift, shift of authority from humans to algorithms,
might reach a point when most humans are simply incapable of understanding what is
happening in the world. Even most governments and heads of state will not be able to really understand what is happening in the world. And more and more decisions will be taken on our behalf by algorithms which is why the question
who designs these algorithms and on what ethical basis
is extremely crucial. So I hope that the
discussion we have today in the coming hour or
so will help not just to enlighten us about these issues but to really spark a public conversation and a political conversation about this. Because again as maybe the last remark before we really begin the debate, to take a long-term
historical perspective, there is always a connection between technology and politics. Technology often defines what are the main political issues of the day. In ancient times the most
important asset in the world was land and the most
sophisticated technologies were the agricultural technologies which was a basis for
agricultural societies. So politics was largely a
struggle to control land. Then with the Industrial Revolution the most important asset
in the economy changed from land to machines and factories. And politics, over the last two centuries, increasingly became a struggle to control the machines, the factories, the industry. Now data is replacing machines as the most important asset in the world. And the main political struggle is no longer about
machines, it's about data. Those who control the
flow of data in the world control the future not just of humanity but maybe the future of life itself. So I hope that this
debate we are having today will help ignite or continue to ignite the public interest and the public debate about these issues. Thank you. (crowd applause) - So, Professor Harari,
please have a seat. - Thank you.
- Thank you very much. We'll be spending those
our and a half together with all our guests. I have to tell you first of all, I'm a very optimistic person and I really believe in
the resilience of humans. But after I read "Sapiens",
I felt a little bit desperate I would say because sapiens
are the greatest predators of all time. Am I right to be afraid? - No, they are not the greatest predators. I would say they are serial killers. Lions and bears and sharks are much better predators than us. But when it comes to
exterminating entire species and entire habitats, there is
nobody like us in the world. Even before the Agricultural Revolution, even before the first
wheat field was planted and the first city was established, humans were already responsible for the extinction of about half the large terrestrial
mammals of the planet. So, yes, there is a
great cause for concern. We should really acknowledge that the future of most
of life now depends on us and on the choices that we make. - So we invent and we have
made great inventions, but also inventions that
have a terrible impact today. We have invented electricity, we have invented nuclear fission, and a lot of inventions that
we are paying the price today. Are we really that
short-sighted as sapiens? Can we not see longer than the
near return on the interest? - Usually not. Some people as individuals are
able to look more long-term, but human institutions have a greater, much greater difficulty in that. And throughout human history, humans were always very, very
good in inventing new things and manipulating the environment, the rivers, the forest,
the animals, other people. But they always had a big problem foreseeing what the full consequences of their actions will be. A great example is the
Agricultural Revolution which the domestication of wheat and rice and chickens and cows and so forth. And when it happened,
lots of people thought this is a great thing for humanity, but actually for most people, life after the Agricultural Revolution was much harder than
before for a small elite, the kings, the priests, the emperors, they had a very good life afterwards. But if you were a simple
peasant in ancient Egypt or in medieval China,
your life was actually much harder than before the
Agricultural Revolution. But it was extremely difficult to foresee the full consequences of that. And it's the same with the
technological inventions of today that nobody knows what
will they actually do to human society and even
to the human body and mind in the next say 30 or 50 years. - But is there a way to, I mean, at least if we don't
know what's gonna happen with all the data that we have
with artificial intelligence, predictive systems, will it help to take the good decisions? - I hope so but it plays
both sides of the game because when you have more data and more computing power
and more predictive power, then you think you can
see further to the future but actually exactly the same technology mean that the change is accelerating. So there is faster and faster change so it's actually harder and not easier to predict the future. It was much, if we sit
here a thousand years ago in 1019, it would have
been far, far easier to predict the future than it is today. If we'd sit here a thousand years ago I couldn't of course predict the political situation in 1050. But I could tell you with great certainty what the job market would look like, what the economy would look like, 90-something percent of
people will still be peasants. And so if you think about what
to teach young people today, a good bet would have been
teach them how to herd cows and make cheese and plant
rice and grind corn. This will still be useful in 1050. Now in 2019 you look to 2050 nobody has any idea what the
job market would look like and what kind of skills people will need. We have a lot more predictive power than a thousand years ago, but exactly because of that
the change is far more rapid. And that's the paradox
in a way of prediction and the paradox of knowledge: the more you have of it actually the more ignorant
you become about the future. - So then how can we be more conscious and how can we act more consciously? We're scared, nobody
knows what's gonna happen. In fact nobody can predict as you said. - Yeah, I think it means
a couple of things. It means that you have to hedge your bets in terms of education. It means for example
don't teach young people a particular skill, don't
place all your bets on the idea that this is what people
will need in 2050. They will need to code computers, so let's educate them to do
that because maybe by 2050 actually AI will be able
to code better than humans so this will not be necessary. So it's a far safer bet
to teach young people resilience and mental flexibility because we don't really
know what kind of skills they will need to learn. We do know that they would
they will have to change. And similarly when it comes to
political and social systems I think the key factor is,
or the two key factors, is balance in cooperation. It would be extremely dangerous if too much power and too much data, and data now is deemed
the raw material of power. If too much data is
concentrated in few places, either in a few corporations
or a few governments or a few countries, we need
to, as much as possible, disperse power and data to more locations. And secondly we need,
again, another safe bet is that we will need global cooperation. The only way to prevent
the worst-case scenarios is if enough countries cooperate on that. If not, what you get is an arms race, and over the last few years, especially the last two years, we have seen an accelerating arms race in AI between the US and China and a couple of other
countries now joining in. And we are likely to
see a similar arms race in biotech soon enough. And if this happens, it means that it will be extremely difficult to prevent the worst-case scenarios because even if you explain
to people the dangers, everybody will say, well,
we are the good guys. We don't want to do this dangerous thing but we can't trust our
rivals not to do it, so we must do it first. That's the logic of the arms race. And if we enter and we are entering a technological arms
race in fields like AI, this is the worst news
possible at the moment. - We're going to talk about
that later with Ken Roth. But coming back to data, the fact that we are losing our privacy, our freedom every day, and we're giving our data
away everywhere that we go, how can we stay free? - First of all we maybe need
to realize that we are not as, we were never as free
as we thought we were. There is increasing
evidence that if by free you mean freewill, then
this was always an illusion. Now in the past you didn't have to pay a high price for
believing in this illusion because humans were too complicated to be hacked by external systems. So even a thousand years
ago many of your desires and decisions did not reflect freewill. They reflected all kinds of biological and cultural factors. But you could still believe in freewill because nobody could really understand how you make decisions and nobody could manipulate people on scale. Now freewill might become
one of the most dangerous illusions in the world because the easiest people to manipulate are the people who believe in freewill because they don't believe that
anybody can manipulate them. If my desires, I mean it should be, it's not a very complicated
philosophical issue. Just the next thought
that pops up in your mind or the next desire that
pops up in your mind, where did it come from? Now if you believe in freewill then you say, well, this
just reflects my freedom, and that's the end of the investigation. You lose all curiosity
and also all suspicion about where your desires and
thoughts are coming from. You just answer, they reflect my freedom. But when you really look at it, I didn't choose to think this thought. This desire, where did it come from? If your curiosity is
really aroused by this and you start investigating, you realize that your desire reflects a large number of biological and social and cultural factors which
are not under your control but might be increasingly under the control of somebody else. And again philosophically
this was also always a fascinating issue. But now it shifts from
the realm of philosophy to the realm of engineering and to the realm of politics because now at least some
corporations and governments are gaining the technology to manipulate and control human desire
on a massive scale. So I would say about freedom we should realize freedom
is not something you have. Freedom is something you
need to struggle for. If you start with the
idea, the assumption, that I have freewill,
any desire that pops out in my mind, this is my freewill. Freedom means exercising my desire, then you're already slave
you're only a slave. You're already a slave not
of your biological mechanism, you're already a potential slave of somebody who knows how to
pull the levers of what really causes you to desire one
thing and not another. If you realize it and you
start at least investigating what's really happening there, this is the road to actually fighting and gaining real freedom. - But then if we compare
your definition of freewill to the fact that, okay,
we have a gut feeling and that's maybe what pops up and we suddenly think,
okay, this is what I think. But on the other hand if
you have all the information in your hands to take a decision, why this decision isn't
based on your freewill? - First of all you can never
have all the information, certainly not the amount of information that today are potentially
available to an algorithm. Usually when a human makes a decision, humans make decisions on the basis of just three, four, five data points. An algorithm can make a decision on the basis of thousands
and thousands of data points. - So that could help us to take
the good decision for good? - Yes, but this means that
the authority is shifting from you to the algorithm. To take a concrete example, if I apply to a bank to get a loan, so 50 years ago or 10 years
ago, on some cases even today, my application goes to a banker who goes over some relevant
files and information and takes a decision
usually just two, three, four salient points about me, maybe it could be
something like my history, my credit history, maybe
it could be my race, my gender, my religion. All kinds of salient features
go into this calculation and the human banker takes a decision. Now an algorithm can, and
this happens already today, can go over thousands and
thousands of data points, many of them would seem to
a human utterly irrelevant. And in any case a human
can't really engage with so much information. For instance there are cases today when such an algorithm it checks when you charge your mobile phone
and takes this into account when deciding whether to give you a loan. It takes into account the day of the week and the hour that you applied, all kinds of things that
might seem irrelevant to us, but the algorithm can
find some pattern there and take this into into account. So first of all it's very
difficult for us to compete in the amount of information. And secondly even if we
have all this information, our understanding of
what's actually happening inside our bodies and
brains is extremely limited and is usually shaped more by mythology and theology
than by science and biology. This is why I said
earlier that it's not only about algorithms and AI. AI by itself can't do
much unless it is linked to biology and to biotechnology, because if you're dealing with something that has nothing to do with humans, then, yes, AI without
biology can accomplish a lot. But the moment humans enter the equation, you also need biotechnology. Even for a self-driving
car to go on the road, a self-driving car must be able to understand human emotions. Now we are very close to really the merger of biotech and infotech. We are still not there. I think that the invention
that will really bring these two revolutions together is the biometric sensor that translates biological phenomena into digital data. And once we have ubiquitous,
cheap, biometric sensors then we have the real fusion
of the two revolutions. And this will basically
change the entire world. - I'm still depressed, sorry to say. Please, I would like to
welcome our panelists. Professor Effy Vayena, Ken Roth, and Professor Jacques Dubochet. Please join us on stage to
continue this discussion with Professor Harari. (crowd applause) - Hello. - Good evening, Professor Vayena. You are coming from the ETH in Zurich. You are a digital ethicist, bioethicist, and you are really
specialized in to health and how we use our data related to health and gene sequencing and
all those questions. Ken Roth you are the Director
of Human Rights Watch, a leading NGO that is advocating to defend human rights
everywhere in the world, were former prosecutor. So we're going to also talk
about legislation with you. And Professor Jacques Dubochet, a Nobel Laureate 2017 in chemistry. And he's also, you are an
environmental activist we can say. So we're going to talk
about environment tonight. So, Ken, I would like to start with you. I think the mics are on. To talk about one major
concern that is populism. Populism is on the rise and it has a lot of
impact on our societies and our democracies. What is your vision? I mean your point of view regarding the rise of populism,
the backlash on tech, on also the environment issues, and the fact that technology may be used by those populist authorities will have an impact on human rights. - Well, thanks, Leila. Let me quickly answer your question and then try to relate
it to Yuval's points. I mean populism for me is a phenomenon in which leaders gain power by essentially demonizing
some unpopular minority. Could be gays, it could be migrants, it could be people of a certain religion. And then once in power they
systematically go after the checks and balances on their authority by attacking independent judges, independent journalists,
independent activists, the elements of democracy that ensure that an executive is
accountable to the people. So that's the essence of what populism is, the rise of authoritarians. And you see this in every
place from Viktor Orban to Erdogan in Turkey, Sisi in Egypt, Duterte in the Philippines, Trump. There are plenty of them. Now to relate this to what
Yuval was talking about, and I think that the
technological innovation that has made it easier
for populist to gain and maintain power today is
the emergence of social media. And what that has done is allowed a much more tailored message. I mean Yuval's point is that
this concept of freewill is a bit of an illusion
that we've always been subject to biological influences but also to informational,
to political influences. But in the past, politicians had to work through mediating institutions. They would have a
political party that would get rid of the Trumps and make sure that a more responsible
figure was put forward. Whoever that person was
would have to speak through institutions like the
BBC or the New York Times which would also filter the message, determine what an objective
assessment of the truth was. Social media has broken that down. Anybody can get on social
media and send the word out. They don't need a political party. And they can speak directly to people. They don't need the mediating institution of a media institution. So this is a new form of influence. It allows anybody to have a voice. Whether that voice resonates or not is a product of their
talent and their message. But it is a more dangerous moment. And particularly when the
social media companies, because of their algorithms,
are prioritizing engagement, and because we tend to
engage with the more extreme, the more provocative. This process is fueling the
kind of populist message, the messages of hate and fear that are driving people like Trump because that tends to be more engaging. That's why people tuned
into the Republican primaries three years ago. It was entertaining. And that same phenomenon
on Facebook or on Twitter drives people to these messages. And so we find that the algorithms which are choices, the
social media companies could be prioritizing something else but it's just less lucrative. They can sell ads the more you
spend time on their medium. So they send you the provocative stuff. It tends to fuel this
anti-liberal anti-democratic voice that is gaining prominence
in many parts of the world. - Professor Harari, you met
with Mark Zuckerberg recently. Are you reassured after
your discussion with him? (laughs) - Not really. I think the problem is much bigger than Zuckerberg in person. He and others like him that have built these immense machines,
and now these machines are really independent of
them to a large extent. So even if he personally changes his mind or has a positive vision, the machine is not necessarily
responsive to that vision. And I completely agree that now one of the biggest competitions is the competition for human attention. This becomes a very important resource. Everybody's competing for that. And the easiest way to
grab human attention is by pressing the fear
button or the hate button or the anger button. And because the entire commercial system or business model of
these social media giants is based on grabbing people's attention, they are almost forced to
do these kinds of things. And so I think it's much, much bigger than any individual person
and his or her vision. And we need to really
change the basic model of the social media industry
and of the tech industry. I mean the really annoying thing is that there is actually so, it's not a lot of money. They don't make so much money
out of these advertisements. It's amazing to think
what political damage is being done for so little money. - Another major concern, if we can say so, or at least a point of
discussion, is China. Professor Harari, I'm
quoting, maybe it's not right that China is behaving like an adult who is tackling the
issue of climate change and the global agreements
needed to contain it. This is what you said. And on the other hand, can you see China as a country where human
rights are highly at risk? So can we mediate
between your two visions? - Or it's just complicated. I mean a country can be
positive and proactive in dealing with one issue, and being backward and negative
dealing with other issues. - [Leila] Maybe we have a
bias in the way we see China, so maybe you can help us
to see it differently? - Yes, I think that certainly
in the last few years, there has been this rising
hysteria about China, that China is gaining on the West, on the USA, on Europe, in economic terms, in technological terms. And this is fueling the current
mentality of an arms race. And I think that, again, the greatest danger is if we allow this mentality of an
arms race to take root because then it means that
we can do very, very little about regulating the
dangerous technologies. Whatever you warn people about, they will say, yes, we
don't want to do it, but the Chinese are doing
it, so we must do it also. And when you go to the Chinese, the Chinese will say, yes,
we know it's dangerous and we don't want to do it really, but the Americans are doing it and we can't allow them to be ahead of us. Now what really happened I think also in the last two or three
years is that America, the United States, has
voluntarily resigned its position as leader of the world and as leader of the free world. Now you can argue to what extent it really fulfilled this role. But for decades it at
least aspired or pretended or claimed to be the
leader of the free world. And then in 2016 basically
the Americans came in saying we don't want this role anymore. We don't care about the world. We only care about ourselves. We don't see ourselves anymore as leaders. Nobody would follow a leader
whose motto is "me first." And I still find it difficult
to understand why it happened and it could have been just
an historical accident, but it doesn't matter. It already happened. And the rest of the world
and in Europe for example can't wait for decades
despite the upheavals in the American political system, there was a bipartisan
agreement that it doesn't matter who is the party in power. America still at least claims
to be the responsible adult in the world and the
leader of the free world. And now this bipartisan
agreement is broken and the world cannot wait every four years to see who the American
public will elect next time. So the good thing about
it is that it forces other countries in other areas like Europe to take more responsibility and maybe Europe can be a kind of mediator or a balance to try and prevent a full-blown arms race in
AI and other technologies between the USA and China. - Ken, maybe you want to
talk about public privacy? You've just published
something related to that. - Let me talk about China here because, I mean, what Yuval was talking about, what is the dystopia of the
future going to look like. The clearest really way into that dystopia is if you look at China today. And that's because China has become an uninhibited surveillance state. It is a government that faces
no domestic constraints. It's a dictatorship. It has been able to
silence public opinion. There is not huge pushback. And so it is operating in a way that, whether it's the Silicon Valley companies or various intelligence agencies
wish they could operate. And let me give two examples of this, and this will illustrate it. One is Xinjiang, the other
is the social credit index. Xinjiang is the region
in northwestern China, principally where Uyghurs Muslims live. And the Chinese see the Uyghurs Muslims and other Turkic Muslims
there as a security threat. So they have imposed a surveillance state like a sort that exists no
place else in the world today. They have police who run around with little handheld devices. And my organization Human
Rights Watch actually just reversed engineered
one of those devices to see what does it show. And it shows that any police officer can gain access to
basically 11 pages of data about every individual in Xinjiang. And this will show everything from their political reliability
to who they hang out with, who their family members are,
what their blood type is, how much electricity they consume, what happens if their
phone ever goes dark, all these supposedly
suspicious activities. And it'll then indicate is this a person who should be interrogated, somebody who should be
detained, what-have-you. That system has led to one
million Uyghurs Muslims being put in detention for re-education. Now one million out of
a Uyghurs population in Xinjiang of 11 million. So basically approaching
10% of the population, obviously a larger percentage
if you exclude children, are in detention today
because according to this intensive surveillance state,
they were deemed unreliable. They were too religious. They were too critical of
Xi Jinping what-have-you. So that shows you what happens
when you have a government that is determined to
surveil for security purposes with no inhibition whatsoever. Now they're beginning to roll
this out across the country. And the form it's taking nationwide is something called the
social credit index. And the brilliance of
the social credit index is that it's a method of control that doesn't usually
even need imprisonment. Because what it will do
once its operational, and they're moving very
quickly toward this, is that everybody in the
country will have a score. And it will be a score in
part social reliability, do you take the garbage out? Do you keep your yard clean? Are you a nice neighbor? You know, innocuous things. Do you jaywalk? But also political things. Do you criticize the government? Do you hang out with people
who criticize the government? Do you appear on panels like this which would definitely
work against our score? Are you in the audience? And everybody gets a score. And then the government allocates
different social benefits according to your score. Do you get to live in a decent city? There are finite good cities in China. Depends on your score. Do you get to send your
kids to a good school? Do you get a passport? Do you get to travel on the bullet train? Do you get to see the latest movie? These little things that people
want depends on their score. And so suddenly because
people want these things, they're gonna start
adjusting their behavior. And it's an brilliant
method of social control that only in the extreme
cases needs imprisonment and otherwise it just controls
people on the basis of data, the kind of thing that
Yuval is talking about. So this is where we're going
if we don't control it. And we're not going to put
the genie back in the bottle in terms of making it
impossible to collect the data, making it impossible to analyze the data. But we have to develop rules
of privacy of data protection that say that there are certain
things governments can't do or not allowed to do with your data. And it's not gonna start with China. They're going to be outliners. But if we can, and this
is something that Europe could be good at, if Europe can develop a strong system about the limits of big data analysis, the limits of applying
artificial intelligence to your personal data, it would help to counteract this trend that otherwise is there and governments are inevitably going to move towards because this is a dictator's dream. You can control your people
without even imprisonment. Why not? So I think that we see where we're going and it's up to us whether
we're going to prevent us from getting there. - Ken, we are here at the EPFL, but also technology can
be used for good also to prevent violation of human rights. Maybe can you talk a little bit about what you are doing now at Human
Rights Watch, on that side? - All right, just so we're
not all negative views. There actually are good thing
you can use technology for, and I'll make this brief. Human Rights Watch, we we
have what we call researchers, basically investigators
based around the world. And so whenever there's a war, wherever there's repression, we've got people on the
ground who who investigate and report on what's happening. And we shine a spotlight on governments and put pressure on them to change. Now governments have
figured out that one way to stop us is to try to block us from getting to the site of the crime or sometimes it's just too
dangerous to get there. And so we've actually
been working with EPFL to develop a remote sensing capacity. In part it's using satellite imagery. There's a Silicon Valley-based
company called Planet which is basically, they they
take a picture of the world every single day. And they've handed us a
million dollars' worth of imagery which is
basically all we could use. And it enables us to say these
villages were burning down or the Rohingya told us the Myanmar Army burned down these villages. Can we see the pictures? And you can actually watch data by day as the villages are burned down, the bulldozers come in
and clear away the rubble and you completely
corroborate the testimony of the Rohingya refugees. Incredibly powerful stuff. And we've done this in northern Sinai for the counterterrorism operation there. We've done it in northern Iraq. We're doing it now in
northeastern Nigeria. And what's interesting is we're combining the satellite imagery with AI because we now have way more imagery than we can possibly analyze. We got a couple people in Geneva
trying to look at all this. But there's a limit to what humans can do. But we are training computers
to look at massive amounts of imagery in certain hot
spots around the world and to tell us day-to-day did
something significant change so that we can have an analyst look at it or send a researcher on
the ground to see it. And what this has allowed us to do is even if we can't get there, even if government have blocked us, even if it's too dangerous, we can observe a lot of what's happening and call out governments
for their misconduct. - So do you have other
examples of the tech used for good like this
one that you've seen with all the, Yuval, the
question is for you, sorry. Have you other examples
that you are like-- - Yeah, I mean obviously, I mean, I usually focus on the bad side because that's my job as a
historian and social critic. You can count on the
entrepreneurs, on the CEOs, on the engineers to publicize all the wonderful things
that the technology can do. So, yes, I mean to give
one obvious example, if you think about self-driving vehicles. So every year about 1.25 million people are killed in car accidents
and traffic accidents, and most of these accidents
are caused by human error. So self-driving vehicles are likely to save a million people every
year, and that's wonderful. Similarly with the
combination of AI and biotech can provide billions of people with far better and cheaper healthcare than ever before in history. So I mean if there were
no positive potential, it wouldn't be tempting. There is enormous positive potential, but again it's the job of
historians and social critics to shine a light on the dangers. And I would like also to
comment on what you just said that I think it is extremely
dangerous what we are seeing developing in Xinjiang and in
other places around the world. And again I think that this, part of the reason I fear
the arms race mentality is that this will amplify
it and spread it around. I mean if the legitimacy for a government like the Chinese government
to do something like that to its population increases, the more it can present itself as being in a life-and-death struggle
against an external enemy. And then also when you
look at other countries like the US, like Europe,
and other countries around the world, which maybe will not
be able or not even be, or desire to implement such
a program in their country, if they are caught up in an increasingly escalating arms race, and again this mentality
of its us or them, there is no third way, then this will lead to
the continual spread of this kind of technology. And the Europeans will say, yes, it's a very frightening idea, but we cannot allow
ourselves to remain behind. Not just in terms of the technology but also in terms of the economy. If the social credit system
means that Chinese companies, let's say, are far, far
better than European companies in hiring people, in
hiring the right people, and motivating them and
getting the most out of them, then you will see increasing pressure from European corporations on governments telling them you must
allow us to do it here, something like that. Otherwise we'll go bust and
the Chinese will take over. So what do you want? And this is why I think that
the logic of the arms race is so, so dangerous. - Professor Vayena, you
want to add something? And then we go to the augmented humans. - Yes, thank you. I've been thinking about
your point of the arms race kind of concept. And precisely what you're
saying is what's happening. In every conference I go,
and I talk about privacy, people tell me, well, if you have privacy, you won't have good technology. And the issues presented is
a conflict between innovation and progress and a human right, because I go back to the
human right of privacy. So it seems to me very
difficult to get out. I try to convince people
we have to get out of that, let's say, conflict. But it seems to be very difficult. And I find this not only
in the political discourse, I find that in our institutions. I am in a technological
institution myself. I talk to scientist all the time. And I get that division it's
either that or the other. And it fits exactly into what you were saying about arms race. If we don't do it somebody else. If we don't do innovation
then somebody else is gonna do innovation. So I'm trying to see,
and that's my difficult, maybe you have a comment on that actually. How do we get ourselves
out of that binary thinking that it's either/or? How can we finally start thinking that if we do have privacy, if we do respect those basic rights, then that's the kind of innovation we want rather than some vague sense of innovation that's going somewhere that
we don't know where it is. So I'm asking that more than you, but because we're
bringing up the arms race and how to get out of this, do you have an idea of
what to get out of that? - I mean, the basic idea,
and maybe it will happen. Maybe it's inevitable, it will happen, we'll get into the arms race situation and then we'll see what happens. But we are still not fully there. There is still time to, I
think, to reverse course, not a lot of time, but
there is still time. And to realize that actually
people say in China, and in Europe, and the
US, and other other parts of the world have common interests. And everybody is going to be hurt if we enter this kind of
accelerating arms race. And that actually this
idea that we have to do it, otherwise they are being left behind. To give an historical analogy, in the 19th century we could have had the same discussion about, say,
kids working in coal mines. So you would meet with the
head of the big companies of coal in a place like Britain and they will tell you, well, yes, sending kids to coal
mines, we don't like it. But the Germans are doing it. If we don't do it,
we'll go out of business and the Germans will take over. What do you want? And eventually people realized
actually it's better both for the Germans and the British if the kids go to school
not to the coal mine, and it's good even for the economy. Even the economy flourishes
if you have more, for us today it's obvious. But it wasn't obvious at all in 1850 that it's better to
send the kids to school and you won't be left
behind in the economic or military arms race if you do it. And I think it's also the
same with issues like privacy. There are extremely high
costs to pay also economically for completely eroding human privacy. But when you are caught up in this, mannequin battle between good and evil, you feel that you have no choice because the other side is gaining on you. - So let's come to augmented humans. You have talked about your
vision of augmented human, Professor Vayena, are we there yet? Can we talk about augmented humans? - Okay, so I need I
guess a better definition of what do we mean by augment-- - [Leila] Exactly, what is the definition? - Because if those of us
who have contact lenses or wear glasses we're slightly augmented in the sense that we can see better, we wouldn't be able to see better. I myself and many others in
here fought a lot of infections over the years, so we're here today. We boosted our immune system, we killed bacteria, and we are alive. So we did better than we would have done had we been left on our own. And if you look at scale for, it's not long ago that the life expectancy was about 45 years old. Now we live to be in
this country actually, one of the better places to live, we live to be 85, 86. And we did quite well. We like that. So I think in many ways
we've augmented ourselves. Now the question I think
in the augmented human is where is that threshold
where we have a norm and we boost ourselves to reach that norm because we were below? Or where we start to be
going above that norm? And my feeling is that we
keep changing that threshold. Now the bigger question
is how do we determine that move upwards I
think because having now wearing glasses or having
laser surgery or whatever, that's something we agreed is okay to do. Everybody will have it. We're trying to make sure
that everybody has it, not just a few people. But when it comes to other
things that are going beyond what would be the norm, then I think the conversation is becoming complicated. Now in what you were
mentioning, of course, we're taking augmented at
a different kind of level with the convergence between the computational and the biological. And I think one example
that we talked a lot about these days is the genetic
engineering, right? If you're able to intervene to that extent that you change the very core of humans in a way that you want because
we do change over time. But how would it be if we could
change our genetic makeup? And we had examples last year, the babies born in China
where the Chinese scientists actually made that step and
tried to make that change. Is that augmented? Okay, so what he tried to do for example was to produce babies that would be, wouldn't be susceptible,
with genetic editing, they wouldn't be susceptible
to acquiring HIV infection. Okay, we have other
means to do that as well, but let's say is that
the kind of augmentation? We are not sure if that worked. We have so many other risks. So if that's what you mean by augmented, I'm not sure how close we are. But as you said earlier,
I mean if we look forward to 50 years in the future, I think a lot of what
we define even broadly as augmented will probably
be getting closer to. - Would you recommend
strict regulation on that, Professor Harari? - Yeah.
- More glasses for anybody. - On the threshold and then
what should be the limit to put. - Is what we call enhancement
above that, right? - Yeah, I think that the key
issue is what kind of changes you do that allow many people
to reach a certain norm, and what kind of changes you
do that allow a small part of the population to exceed the norm. There are two different projects here. Now there is no clear
technological and scientific line between them because
very often a technology that can help people reach the norm can also help some people exceed it. The key point is how
do we make it available to the masses of the population? Otherwise we end up with biological castes with maybe a small caste
of enhanced superhumans and a massive population
of just normal Homo sapiens which is increasingly left behind. Previously in history you
could not really translate economic inequality into
biological inequality. To some extent yes. Rich people had more food so
they were taller and so forth. But the basic human abilities, there were no real difference between the daughter of the king and the daughter of the peasant. But in 50 years we might reach a point, if we are not careful with
regulation and so forth, when humankind might split into
different biological castes. And this is something, again,
it's a political issue. Scientists should be
aware of the possibility but ultimately it's the
responsibility of governments and citizens to think about it and to prevent it from happening. - You want to add something? - What we know from history
is that technological advances are not equally distributed. We know that. Even if it's not the enhancement, if it's not the extra
thing above the norm, things that would allow
people to reach the norm, not all of us have them. And we have big parts of our
world that don't have them. So if we were to judge by that, I would say it would be were
very difficult to imagine that all these advances would be suddenly equally distributed. I would be very concerned
about that as well. And so the question is if
how much that consideration becomes part of how we
decide for what we are going. So if Project X seems
fantastic, it is brilliant, our curiosity will be met, and we'll be thinking
that we are super humans. Is that the project we're going to pursue provided that that's
an outcome that we know it will benefit only a few? Is that a consideration that we need to be taking into account when
hopefully we make decisions, and not decisions that
are made by somebody else, about which kinds of
innovations we're after and which kinds of technological projects we want to pursue? - We're going to talk a little bit later with Professor Courtine,
Bloch, and Paik about that. Professor Dubochet, you are very concerned about the climate change. You are an environmental activist. Do you think that, and you've been actually
with the kids on the streets to protest against the
non-reactivity of our politicians. So do you think that we, the people, can be agents of change in tackling those environmental issues? - Yes, they can. We are not going to having concurrence your horror with my horror. I take your horror very well. Human animal being at
present already hacked. And the society, the value
I share are in great danger. I agree. I see with you. And I read your three
books, they are very good. But, well, my problem is climate. And since a long time, since seven, and I saw with horror how
little we could do with that. And you see with horror
how little we can do with your horror. And we were thinking and
discussing this yesterday in family and we remember that Al Gore was Vice President of the
United States up until 2001, if I remember correctly. And he was a powerful person. A few years later he brought the film "The Inconvenient Truth". We knew all that and nothing happened. Nothing happened. But I have my heroes now. Great Totenberg in August last year. It changed a lot of things. Or at the same year at the same month, Aurelia Abajo in France. Aurelia Abajo will be here in Lausanne the 3rd of October, great. Now here I see something changing, and these young people are saying, "It doesn't work like that. "We don't accept this." And I have the hope that this movement gets strong enough to save our climate, but they are clear. They say it's not a matter
of changing the climate, changing our society. And they say it very strongly. And I hope that on the
move about the climate they will save us from
becoming slave of hacking. That's my hope. (crowd applause) - A quick question on
that, Professor Harari. Global cooperation is
needed but we also know that it's a very slow process. It takes ages to have
international treaties. So what other solution do you see? The civil society
rescuing the environment? What other solution do you
see like Professor Dubochet? - In terms of the climate,
and, yes, there are many things that individual countries can do. And many of these things are
not even bad for the country. I mean I think we should get
out of this kind of, again, binary thinking that the only
way to stop climate change is to destroy the economy. If this was the case, then
there would be no hope. But I think it is possible
to spread prosperity and at the same time to have far more responsible
environmental policies. But again the problem
is that there is a limit to what individual countries can do. Unless you get at least
a substantial majority of the main countries on board, it will be extremely difficult to prevent the worst
outcomes of climate change and of the ecological crisis. And again this goes back
to the issue of inequality that climate change will
have a very different impact at least in the short-term
on different countries and on different social classes. So some countries and people
are extremely concerned but other countries are not
concerned or even quite happy with the direction that
the climate is taking. And again I might try to
be optimistic about it, but in the last few years the world has been running
in the opposite direction. I mean maybe 2015, if you think
about the Paris agreement, 2015 was really the last year that the Paris Agreement
was even possible. It was like a kind of small miracle that they didn't delay it by one year because then it would have fallen apart. And since 2016 that really
the world has been running in the opposite direction. So I hope we can reverse course in time. And again the main problem is that maybe there is no we when we
talk about climate change. Just as with human enhancement
there is no human we. We are talking about
different human groups with different futures and humankind splitting into different maybe biological classes. Similarly with climate change, part of the problem is that
there is no human collective. There are different groups
with different futures. - So now I suggest that
we go to the last part of our discussions. And please welcome Professor Courtine and Professor Bloch on stage
so that we can talk about their research, what they're doing. Professor Courtine is our specialist into neurorobotics at EPFL. I'm just taking my notes
because I don't wanna say he's a leading expert
in neuro technology. And Professor Bloch is a leading expert in neurosurgery at the CHUV. Thanks for being with us. (crowd applause) - Good evening and I
will start by apologizing because we're going to be
a little bit provocative based on the recent discussions. And I will start by introducing David. David, nine years ago,
had a gymnastic accident. And now when his brain send a
command to activate his muscle the signal is interrupted
leaving him paralyzed. Nevertheless when we
studied this kind of injury in rat model, we always find
some spare nerve connection. You see them here in white. However they are functionally silent meaning that the rat
cannot activate his muscle in order to walk. And for the past 30 years, in order to restore walking, scientists have been trying to grow more of this kind of fibers. We thought about the problem
completely differently. We actually focused on the
region of the spinal cord below the injury that is completely intact but disconnected from the brain, meaning it's missing
key source of modulation and excitation in order to be functional. So we thought to hack the spinal cord. Take advantage of advances in technology in order to reactivate the spinal cord with electrochemical stimulation that will mimic the way the
brain activate the spinal cord. This is a paralyzed rat. Stimulation on mean it's turned on. When it's turned off it can't walk. Back on and the rat walk immediately. - And that was the time when we decided to do exactly the same
therapy but in human being. And my role was to
implant an electrode array on the spinal cord at the region that is controlling the legs and to connect it to a
sophisticated pacemaker that is able to deliver
bursts of stimulation that will be located
exactly at the right place and also at the right
time in order to coincide with the intention of
walking of the patient. So here you have David again. David, at the beginning of his therapy, with the stimulation on. You see that it's turned on he walks. Turned off he cannot walk. We turn it on again and he walks again. So, that was very important. - Yeah, because we can train. - Yes, he could train,
trained intensively, very intensively. And six months later he was
able to turn on his stimulation talking to his watch. - STIM on.
- Okay. Start message sent to implant. - And this allows him to
walk outside the laboratory. - However, this stage David cannot control the stimulation on his own. So we had the idea to connect the thoughts directly with the stimulation. Do you think it is science fiction? - Yeah. And here we hacked the brain finally. So here I implanted
electrodes in the brain region that is controlling the leg movements. - And with this electrode we
can record brain activity, build algorithm that detect
the intention of the animal. And we linked them wirelessly to the spinal cord stimulation system. So this is a model of transient paralysis. The right leg cannot move. Without any training, the
animal just think about walking and we connect digital
bridge to the stimulation, and it works continuously
just thinking about it. As long as it's on, he can walk. We turn off this digital
bridge and he cannot walk. Let's dream with me. Let's project ourselves 50 years from now. Imagine someone, the computer-- - Turned off.
- Turned off. This is when the
technology is not helping. Can you somehow restart the video? Or I just make magic mimicking. Yeah. - [Leila] Someone hacked the system. - Imagine a future when
this kind of technology can go to humans. Someone will have the spinal
cord simulation system. We will also need more connection, so we will have gene therapy, all the stem cell therapy,
to grow more connections. And maybe, oh no, it works. So this is the vision of
the future 2050 as you said. The spinal cord stimulation system growing more nerve connections, so
stem cell, gene therapy, because we need more nerve fibers. The same surgery you can have this implant to record brain activity,
decode intention, and linking them to this implanted device so that people can start
training intensively, actively. And the technology becomes
seamlessly integrated in the operation of the
central nervous system to the point that David
can now walk freely to his office drinking his coffee. I know at this stage (crowd applause) it is still a dream. But 15 years of research took us one step closer to this dream. The steps taken by David and six other previously
paralyzed participants. And now we ask you, is it a progress for Humanity? Should we continue in this direction? Thank you very much. - Thank you. (crowd applause) - So, Professor Harari,
what impact do you see at this human-machine
interface for humanity? - The intention of this
research is amazing and positive and can bring so much joy and
happiness to so many people. On the other hand what
struck me about the research is that for it to work you
need to decode the intentions of the brain which is exactly the issue of hacking human beings. So what do you do when you can hack, an external system can hack the intentions of the human brain? And so think about the
politician you most dislike today in the world and ask yourself what would he or she do if
they had this technology to decode human intention reliably on a mass scale from outside? Again it doesn't mean that
we should stop all research in this area. We shouldn't and we won't anyway. It does mean that we need to understand the immense political stakes that such technology involves. And also I think it
means that the scientists who are working in these fields, part of their responsibility
is to educate the public, first of all to educate themselves about the social and
political implications of what they are doing. Very often when you are in the laboratory and you're working on
a particular project, then you only see the immediate project of the technical problems
that you need to solve. So part of it is thinking
much more broadly what are the implications
of what I'm discovering and developing not for
the particular problem that I want to solve, like how to enable
paralyzed patients to walk. But what will this technology
mean on a greater scale for society and for the
political situation? And then to educate the
public about these issues so that politics doesn't
lag too far behind the technological advances? - But there's a gap now. Maybe, Professor Vayena,
you can tell us more? There's a gap between the
pace at which technology is evolving, and it's evolving very fast, and our response as
politicians, legislations. There's a huge gap. And this gap will probably
never be gathered. So how can we do, I mean
the work you're doing, Professor Vayena, what can
you tell us about that? - So, anyway, fascinating technology. And congratulations. That's about reaching
the threshold of norm rather than enhancing. So I don't think there is
something to say against that. But it's absolutely true
that we have set a fast pace in technology that our legal systems and also our societal
norms cannot catch up. I mean it's quite difficult. What we're seeing at the
same time is that even if we cannot change instantly our laws, we have other softer means with which we can decide
where we want to go. And I think I have to
say that although laws are the ones that are
going to, they have teeth, they can make us do things that otherwise wouldn't be able to do. I think this conversation
we're having around what are our soft law tools that can be a little
faster in development, even if they don't have teeth? I do have hopefully not naively
quite a lot of hope in that. And we see also in the
scientific community, in engineering communities, not just out there in the humanities, people interested in
asking those questions and wanting to set a
certain ethical standard. So I think while we're waiting for our legal response to arrive, I think these other tools are useful. And we have frameworks
and systems that allow us I think to bridge, at least
to some extent, that gap. The challenging question
would be how do we do that at a global level? How do we do that with of course the political situations
and the different drives that we see in different continents and in different countries? But I have to say that looking at how, take the biologists and
gene editing community, all of those, I think they are, there's a lot of I would
say sincere interest in developing the standard
because it's not clear what the standard is. It's not that we got the right question, the right answer to should we do X or Y. That is something we're debating. But I think the first thing is that debate to have it, to ask the
question, to have the debate, and set the standard at
the soft law still stage before we get more stronger
response from regulation. - Ken, you want to add something? - Well, I agree with Effy that it's, laws are really just one tool. To take an example that
we're all familiar with, today there's an expectation
that when you buy a product that it will be produced
with a clean supply chain. That even if there's two
or three layers between sort of miscellaneous sub
suppliers in your final product, that you're not going to
find child labor there, you're not gonna find forced labor, you're not gonna find
overt discrimination. And the truth is there are
no laws that dictate that. But rather public expectation
has come to demand that. And it's the possibility of a media expose or civic activism that
has driven corporations to now abide by standards
to go beyond the law. And those that don't are
really dumb and short-sighted and they'll get blown up
at some stage in the media. So that is an example of the importance of the public setting standards. Now at the national
level laws still matter. I mean some of the stuff
Yuval was talking about, if you go into the bank for a loan, the laws already say that
there's certain things that the bank can't do. They can't say, oh, you're Muslim. You're not going to get a loan. There are laws against that. So we need to enhance those national laws and that's easier to do than
to get global standards. But because of this arms race
that Yuval was talking about we need global standards. And I think we've learned a
bit about how to get there and how not to get there. The way not to get there
is to insist on unanimity before you have a standard. And the example I'll give is land mines which my organization Human Rights Watch was very involved in
pushing for the treaty on. And we actually shared
in the Nobel Peace Prize when we got it. We started off down the road in Geneva where the rules at the UN were unanimity. And needless to say many governments, including at that stage
the US and the Chinese and the Russians, didn't want
a treaty banning land mines. So they were gonna just
talk and talk and talk and it would never happen. So we went to a few
countries, Canada and Norway, and said why don't you host a meeting where people who want a treaty can come and people who don't can stay in Geneva? And so they did that. And there was the Oslo process where a group of people agreed on
a treaty banning land mines. Said leave the Americans and
the Russians and the Chinese talking in Geneva, who cares. We got the treaty. And the treaty, although it
was not universally ratified, we'd say it first ratified by
a hundred or so governments. It became so powerful in terms of setting public expectations
and stigmatizing those who didn't follow the rules that even though the
Chinese and the Americans and the Russians still
haven't ratified the treaty, they don't dare use landmines. And the same thing has
happened with child soldiers, with cluster munitions. We're trying to do it right
now with killer robots, fully autonomous lethal weapons. The key is to start with
the coalition of the willing and then expand out. Now what makes this
particular topic so difficult is that landmines we all know
whether they're used or not they go kaboom, it's easy. You can see it. But what do you do if you're developing a hyped up surveillance system? To some extent we can know
that that's happening. But a lot of it is hidden in
code and hidden in backrooms and it's really hard to know. And so I think the challenge that we face is how do you develop norms
when we don't necessarily know what governments are doing. I think it's still possible. It's going to depend on
still the possibility of local exposes, of pointing
out where governments transcend the norms that at
least most of the governments of the world have endorsed, and then making them
pay a reputational price for that failure to abide. I think that's the way forward
in terms of global standards, not unanimity. - Just a quick thing. I think, yeah. (crowd applause) Yes to that. It's not just the government though. I think the powerful corporations, they need to be subject
to similar scrutiny because at the current moment we have to look at those as well. And what we see in terms of
what also Yuval was saying in terms of data control, flow of data, and therefore the
development of technology is exclusively almost in the
hands of some corporations. So their responsibilities
are important as well and they have to go--
- I totally agree. And corporations are very
susceptible to public pressure beginning with their employees who don't want to do this stuff. - But one of the solutions
shouldn't be decentralization if we are scared of being hacked? We could give that to an independent body, decentralized body, and
protect it from states or governments and maybe go more for a completely independent
system from the state? - [Yuval] To give what to an independent-- - To avoid being hacked as humans, and if we give all our data
and all our very personal, I mean, data to independent
body that we can trust, that is not private, that is not public, but that is maybe an independent body, private-public sector, is it something like
decentralization could help to protect individuals? - In principle the more actors you have, the more checks and balances you can have. But giving all the data
to one independent body, I mean I don't think you
can, who will be this body? Who will decide what it will do? It's the most important
resource in the world. It has to reflect somehow
the desires and the values of society so it cannot
be completely independent of the government. And in any case, even if
you give all your data to one body, it doesn't prevent other institutions and organizations. I mean that the data,
and it's not like land. With land, if you have a piece of land and you give it to somebody, that's it. That person or that body owns that land. But with data it can be
replicated any number of times. So I can give my medical
records to somebody I really, really trust. It will not prevent all
kinds of corporations and governments from
trying to get the same or even more medical records either by hacking that system or by independent means of surveillance. And with the rapid advance
in the technologies of surveillance, it's going
to be almost impossible. Even today individuals
have very little idea what kind of data is
being gathered on them every moment without their knowledge. And even if they don't have a smart phone or an email account or whatever. - So now I would like
to invite Professor Paik to join us on stage. Professor Paik will talk
about the latest research that she's done that has been just published today in nature. Professor Paik is a soft robotic expert. Thank you for joining us. (crowd applause) - Good evening. I want to talk about the
future of technology, the immediate future. And I want to call this recovery robots. It's not necessarily
because my lab is called Recovery Robots Lab but it really is. So when I was studying I thought the future robotics
would be something like this. Now we are talking about killer robots and davides, so maybe this is
not the best image to show. But the movie had a huge impact. I did not necessarily want to
create the next Terminator. But it was more to understand what is actually possible to do in terms of physical
material and design wise to create systems that
can mimic human movements. So the eyes that move more like humans, and hands that move like humans. Again not necessarily to
create the next killer robot, but to understand the
biomechanics of humans and to understand what can we do to better make the devices
to work with humans. So one of the last studies,
while I was making that robot, was to create a human arm, a humanoid arm. It actually was able to
create a very smooth motion, enough momentum to even hit a tripod. So that's why the film is
actually shifting at the end. This is really nice. But as soon as someone sees this, you're like, oh great, Jamie. When can they start serving
me breakfast or doing dishes and doing my homework? None. Zero. Even though motion is very similar it does not have enough
intelligence to understand what you want it to do. So I realized this is not
the future of robotics nor the technology. The next generation technology and robots that would actually help us will be a system that's more interactive. Not necessarily it's using
the data you're giving it but because it realizes the environment, That required a huge change in the paradigm of design robots. It sounds really complicated but it's not. You've all done this. Origamis. Origami is a really
versatile design platform. You take a piece of paper, and depending on how you
fold it and when you change the sequence of folding, you can create multiples of animals, objects and systems, or
even just simple rhinoceros. We took an idea from
this origami platform. Imagine having a robot
that you don't really know what he needs to do immediately because that's how the
classical robots are designed. You know exactly what the task is. You want that robot to perform that task better than humans, faster than a human, more accurate than human. But if you don't know
what you want them to do, what kind of robot should you have? You rely on your additional help or another manual type of labor. So you've worked on that idea. How about making a robot
that does not know exactly what it needs to do but
interact with the environment? It reconfigures not only its
body from a flat sheet of robot to a self morph into a
three-dimensional robot that in this case crawls? It goes from point A to point B. But what if the environment changes? What if it's no longer flat? What can it do? Normally the mission
does not go on anymore because it's a different ground. But for Tribot, by it's
able to roll it over. Now you're attacked by the rough terrain. What if it's no longer
just a rough terrain? What if it meets an obstacle? It will hop over. Again it's interacting
with the environment. The same hardware, same controller, but it's able to interact
with the environment. And it can even do gymnastics. It's a very simple example, but this is a valid objective robot. It takes a sensory
information, it process these. But because it's able
to reconfigure its body, it's able to create
different gait patterns. That's not it. It does not necessarily have
to be all the same flat sheet. We can make them
completely modular as well. So instead of relying on a
single sheet with a fixed size, we can create multiple modules that are either active with motors and sensors and microprocessors, combined with inactive or passive joints. In this case you see a triangle. And by combining them like Lego blocks, you can create not only
different size of elements but a different platform
that can interact with you and the environment. So does that sound like something that you'll be interested in using? Maybe not immediately because
you never seen it before so you don't know what to use it for. But there's a place that you really need this type of technology. Imagine if you have a robot that needs to go through an obstacle. In this case we're showing
it by showing a wall with a small opening. And the robot like this
can transform its body to execute its mission,
in this case turning it to a space shuttle. And speaking of space shuttle, space is the place where you
need this type of solution. This is an artistic rendition
of you're talking about. To bring up one kilogram of anything, let alone water and food, it would cost you
hundred thousand dollars. You can no longer afford to
have automated system of robots for a single task. You need a system that
will do multitasking. This is where Robogamis can come in. Imagine having multiple modules that you don't know exactly
what missions they need to do but they're gonna be there
to help out astronauts. In this case instead of sending
astronauts out in the field, these are the robots that
will go inside the ground, survey on top of the ground,
and even above the ground. Not only that they will
interact closely with astronauts by providing a platform
that is interactive, physically interactive, with astronauts to communicate back with the Earth, helping you out with experimentation
on the space shuttle, or space ship. This is not the distant future because you can use a
same origami platform that transforms its body from flaccid to a three degrees of freedom force feedback haptic interface. What you're seeing here, it's world's smallest haptic interface that gives you force feedback
underneath your fingertip. That's very mouthful I get it. What does it mean by force feedback underneath your fingertip? This is an interface. If you were to link it
back with your goggles, or virtual reality goggles, normally when you play with
the virtual reality system, what you see is what you get. Well, actually you don't get it because you probably wave
your hand in the air. But if you combine this
with this haptic interface, which you're seeing just
underneath the thumb there, these are no longer just
blue ball that he's touching. It's blue rubber ball because you'll be able to
feel the stiffness change. Now it's not just a red
ball, it's a red sponge ball. As soon as you pick up the black ball, you'll be able to feel
that it's a billiard ball 'cause ceramics are much harder. Imagine using this interface
to perform surgery. Imagine using this as a layman person buying your avocado. You would know how hard that is. And since we're in Switzerland. (speaks in foreign language) It's reality now because
even though it does not look like your killer robots, these are robots that can
enhance your experience online. So people often ask me, Jamie, those are really cool-looking. But what's your main application? This is a wrong question now. You're living in the next-generation
technology driven world where the robots are no longer designed just for a single task. They are no longer optimized
for a single task anymore. I really think the next-generation robots are meant to multitask. They're optimized for multitasking. That are highly interactive
that will bleed into our lives not to take over. We are able to keep our autonomy, but these will interact with us to maintain the quality
of life that you desire. Thank You. (crowd applause) - Thank you very much. So, Professor Harari, would
you like to be surrounded by mini robots, origami
robots, like these ones? - Depends what they do. I mean, again, like every technology it can be used for so
many different purposes. But what strikes me, again,
like in the previous examples is that the really crucial point is not the robot by itself but the ability of the robot
to interact with humans in order to change the human experience. So for example to feel through the robot, what some robotic arm is
doing in another place. So this again it bridges the
gap between the human mind and the human experience
of subjective experience and the objective world outside. And once this gap is is breached, this is when the biotech
and the infotech revolution really combine to
completely change the world. - We're slowly coming to an end. Obviously you're on a mission. So what are the next steps for you now that you have the latest
book that has been published? What are the next steps? - The really key issue is to change the public conversation, to focus the global conversation on the most important challenges that are facing humanity. I see myself as a kind of bridge between the scientific community
and the general public. And so I my next, my
team's next big projects is really how to widen the conversation. So write to children about it, write a children's book, create maybe a Hollywood
blockbuster about these issues, because I don't think
that it should remain an internal conversation of
the scientific community. And these are the best ways
to reach the general public. Maybe the main message to
the scientific community is that just as the border
between biotech and infotech is disappearing, so also whatever remains of the border between science and politics is also disappearing. And most scientists don't
like to hear this message. Most scientists, they really
want to do just science. They don't want to get involved in the messy issue of politics. But in the 21st century
science is the most important change factor in the world
in the economy, in society. So it is also the most
important political factor. So on the one hand politicians and voters should be far more educated
about what's happening at the the front lines of science, and at the same time scientists
should be far more aware of the implications of what they do for the political system and for society and should take greater
responsibility for that. - Thanks a lot, Professor
Harari, Professor Vayena, Ken Roth, Professor Dubochet. It was a great honor to
spend this evening together. It was enlightening,
so really thanks a lot. And I would just close this panel, so if you wanna go back to your seats, you are free to do so. I would like to thank
our supporters tonight because this event, free event, that has been offered to you has been, can we see the sides
with all our sponsors? I would like to thank the two, three main sponsors of this evening. We have Banque Landolt, we have Sigma, and we have Frontiers, Kamila Markram. Thanks a lot for your
support because without you this event would not have been possible. I would like to thank
also the Ecole Nouvelle. We talked about a new education system. The Ecole Nouvelle was
here, part of our sponsors. I'm a proud alumni of this school. And I really thank you for
having supported tonight to welcome Professor Harari. And we have also, (clears throat) sorry, Nomads Foundation, the
beautiful CXIO Foundation who are with us. And of course the (mumbles) Foundation, Stephanie and Joseph (mumbles). Thanks a lot for your support. The Bank Julius Baer. (speaks in foreign language) Thanks a lot because
really it means a lot. We are on a mission. We have also on this mission
our logistical partners, Hotel Royal Savoy in Lausanne. Lausanne Limousine and force. So on this mission if you
want to be part of it, please go to the next slide, and also on the website mfound.org, we need you to come with us and to be also empowered minds because we've discussed tonight if the debate is not among us, nobody's going to lead it. So if you want to be also a member, if you want to participate in
other events like this one, please join us. You go on the website and you can decide to become an empire member. Professor VanDerGinst,
please come back to me. We are reaching the end of this evening. I hope you had a great time all together. Professor, I leave you the stage. Thanks a lot. (crowd applause) - Well we promised an evening
of questions and surely there were many, many
interesting questions today. There were a few answers
as well and some trends. Leila, you started by
confessing some lack of optimism at the very beginning of the evening. And surely there's a little bit of that. And I think maybe in one minute, if I have to summarize
everything we said today, it's starting with that, the fact that technology
is advancing at such a pace these days that, excuse me the pun, we certainly collectively
feel disempowered sometimes. And what we really need to
re-establish is this feeling that in some sense at least we are in command. And there's two ways
that, well, several ways, but at least two that
were proposed to us today. One I certainly took note
for myself is education. And not just education in the algorithm. We need to put people in
command of these tools, so we need to educate
outside of universities reach out to people. You were saying, Professor
Harari, that you would go and make a blockbuster at
Hollywood or maybe a kid's book. I think that certainly great
tools and great endeavors, we have to reach out
to more and more people to explain them the
sense of this technology, what we can do, what we can't do, and certainly what we
have to pay attention to. And a second aspect was this
idea of global cooperation. More than ever we need that
because not a single country can tackle these issues. And the last thing, and
that's interesting maybe to mention it, as a closing statement on the campus of a
University of Technology, was this idea that
humanities form a bridge between technology and the people. And I think certainly
here on the campus of EPFL we have to think more and more about that. We master technology. But maybe we also have to be partner in constructing that bridge
that brings technology and empowerment to all of us. Thank you for being here this evening and we hope to welcome you again at EPFL. There will be many, many more celebrations for our 50th birthday. But certainly this was a very good one. Thank you. (crowd applause) ♪ It was a little bit shady ♪ ♪ Always tryin' to play me ♪
