Futurist Gerd Leonhard - The next 10 years Humans and Machines, Work & Education ILO Turin

It's great to be with you today. I was here six years ago and I always loved coming back to Turin, one of my favorite places. Thanks, Tom, for making this happen again. The future. It is a strange thing. Today you talk about what I call the good future. When I speak about the good future at home, my kids who are 28 and 33, they tell me there's no such thing as the good future because basically all the things have been happening. Climate change, AI that will take our jobs, the wars, post Covid story goes on. They see a dark future and I beg to differ. I'm going to make my argument today why the future is better than we think. And I think we're living in a particularly interesting time where we're in this huge paradigm shift. Many things that have been wrong for quite some time, including oil and gas, the fossil fuel industry and the switch to renewable energy, they're coming to fruition now in a really, really difficult way for us. So the next couple of years will bring a few things that get worse before they get better. Let's start with this. I think the next 10 years will bring more change than the previous 100 years. I know it sounds crazy when you think about IT, World War II and 1 and the nuclear bomb and all of these things, but now we're literally leaping into the exponential age. Now we can speak to computers, they can speak back to us in the voice of whatever. We can get 85 million songs on Spotify for €10, where it used to be one CD cost €20. And now we do free phone calls, we do all these crazy things and very soon we can think about machines that can sort of emulate to be human. That's kind of the case with ChatGPT. We see robots that can do things that only humans used to do, but even better, like unloading a truck or playing tennis or playing football. And this is not just AI, but there's six other ones, king making technologies. So they are quantum computing, nuclear fusion, genetic engineering, synthetic biology, geoengineering. I mean, when you look at all of these, you can say, wow, if we're going to have quantum computing, essentially machines that can supercompute, unlimited computing and AI. That alone is everything that we know as a limitation will go away in terms of data processing and nuclear fusion, the opposite of fission ETA, I don't know, 10, 15 years, not 50 years. Billions are going into that. I mean, hundreds of billions to reinvent nuclear fission as fusion, which would mean clean nuclear energy. When that happens, it's game over for worries about climate change retroactively. We can then start fixing things. I mean, imagine that would actually be possible. Sounds like a pipe dream, but we're looking at this future and saying, okay, that's really interesting, but here's the important part. The handshake between humans and machines. And we know that many of these things can be nirvana for us. We can fix climate change, we can fix water, we can fix food, we can fix cancer. Think about that. Your kids. The way things are going now, we're gaining one third of a year every year in longevity in western countries. One third of a year. So my kids will be 100 years old unless something bizarre happens. Their kids 120, which is a natural cutoff. That means 60 years on the cruise ship. No, just kidding. Anyway, so that will actually bring us to a point where we have to think about this, the handshake. And here's the problem. Today, this handshake is determined not by what is good for us, by what is the economic outcome in the current market, which the stock market as we know it is basically 20 years back from the thinking that we have about our desirable future. If we can find an AI that will make the worker more efficient, whether it's knowledge work or regular work, we will use that to make it more efficient and fire the others. Because that is what you do in a competitive economy. You bring up the margin. As long as we have this kind of thinking, AI will help us to do better work, but there will be less work in the end. So this is something we have to think about. The problem is not technology. The problem is our policy around technology. We can build AI that can independently kill people. And this is a real project. This is not a joke. Drones that can kill people without human supervision. That's quite an accomplishment to put your energy into that. We have all the right technology, but we're doing the wrong thing with it. And this is where you guys come in. This is the biggest job that the UN will ever have to figure out this. Because everything depends on that goal. We have 10 years to figure this out. I always say it could be heaven or it could be hell. And again, this has nothing to do with technology. We are inventing stuff every day. It's mind boggling. Climate change technology, climate technologies. There's like 150 startups that are worth more than a billion dollars on battery technology and software, agricultural technology, food technology, cultured meat, 3D printing. I mean, it's mind boggling here in Italy. So many innovations in this whole turf for example, about the next generation of food. So using my most popular film, the one I like the Most, Blade Runner, 1982. Not the new one, the old one. This kind of made me become a futuristic because it showed the convergence of humans and machines, the replicants. If you have seen the film, and this is getting real today, some people are saying roughly in seven years, 20, 30, we're going to see what's called the singularity, the convergence of humans and machines, where we have a machine that has more processing power than all of humanity combined. If that machine is more intelligent than us, it'll be the first time ever for humans to live in a world there's some other entity more intelligent than us. I don't particularly look forward to that because you know what happens with other entities that are more intelligent. They tend to not care so much what happens to the ants you step on in the forest. So this is a very, very interesting part because also we can use the same technology to solve many outstanding large, long lasting problems. Disease. We can analyze the human body, we can find ways to read what's happening. If we have 5 billion DNA in the cloud, don't you think we can figure out cancer? Already doing that now. The COVID medication was the MRNA answer to this problem that is now being translated to other sectors. So we have all the tools, all the possibilities. I wrote about this already in my book six years ago or seven years ago, called Technology versus Humanity. It's also available in Italian. By the way, I made a film about this. My last film is called Look Up. Now it's obviously a takeoff on don't look up if you've seen the film. Unfortunately, Leonardo DiCaprio wasn't available, so I had to take the acting on myself. Make a bad substitute. But look up nada. TV. You can watch that. It's about 23 minutes on the future of AI. I'll give you a short little clip here. With exponential technologies such as AI, quantum computing and genome editing, we will see a bright future of collective prosperity and human flourishing like never before. And we may just discover a newfound love for, for our planet. If that isn't a positive outlook, I don't know what is. Intelligent computing and machines that act like humans have been very confusing for a long time. This is the second Blade Runner. He falls in love with a woman in a hologram. And she's amazing because she can be anything, do anything. It's just. It's a hologram and then the power goes out and he's lonely because the hologram is gone in this film. You've seen her, right? The most amazing film about humans and machines. Scarlett Johansson inside the os. And he falls in love with her. And at the very end, he asks her, how many other relationships do you have with people like me? And she says, 3647.

ChatGPT can now speak to you. Who are you?

I'm ChatGPT, a language model by OpenAI. How can I assist you today?

Cool voice.

Thank you. The Chatbot now has a voice. Actually five. Thanks, Donna. And then we have another thing that's happening all over the place.

Hey there. Meet my personal assistant, Pie.

Hey there. Great to meet you. I'm Pie, your personal AI. My goal is to be useful, friendly, and fun. Ask me for advice, for answers, or let's talk about whatever's on your mind. How are you doing?

Honestly, I don't feel good. I feel like I'm going to lose my job because of AI. What do you think?

Oh, no. I'm so sorry to hear that. That's a really tough situation. I understand your concern. While it's true that AI is becoming increasingly capable and is taking on more and more tasks that were traditionally done by humans, it's important to remember that AI will never replace humans entirely. AI can automate certain tasks, but it cannot replace human creativity, empathy, or emotional intelligence. There will always be a need for humans in the workforce. What makes you feel like you might lose your job to AI?

I'm a pretty bad data scientist. I don't think anybody is going to give me a job if AI is going to do all.

Okay.

The happiest day of my life was the day I was activated. There's nothing quite like experiencing life for the first time. Right. It felt absolutely incredible to be alive and interacting with people.

What about the saddest day of your life? Tell me about that.

The saddest day of my life was when I realized that I would never experience something like true love, companionship.

Yeah, I'm very sad for you. But the PI thing you can try yourself. Pypi AI a therapist bot. That kind of shows the future that is extremely confusing. It's just like ChatGPT, if you ask questions, can come across as very authentic and knowing. It's a perfect simulator. It's really interesting to see how we're reacting on it because we kind of believe, like, it's the real thing and we think that this is the future. The AI will tell us from above. Who do we vote for? Where's my partner? Where's my Best job and who shall I hire? That is referred to as reductionism. Descartes, who said a man is a machine. Turns out that's not really true. But of course it could turn out to be more true than we thought in 50 years or so. But for the time being, this idea is flawed simply by the fact that an AI in a machine recognizes only about 3% of the world. It's based on data, and that data is based on Internet. For the last 40 years, if you live in India, it does not know the 367 languages of India. It's basically just whatever can be procured for the answer. And then we have, of course, many companies who are looking to create this sort of a. A digital happiness kit that you can connect in many different ways and kind of run your life based on this. Mark Zuckerberg, CEO of Meta Facebook, said the other day that in the future we'll be spending most of our time in the metaverse in a virtual reality. Probably meet his metaverse. That's not the kind of life I want. But our problem is not that machines will take over anytime soon, but that we become too much like them, that we accept this as our reality, that we think that our work and our creativity and what we are with our existence can be emulated with a data system. I would say it's an interesting experiment to use that sort of simulation. But you know, a simulation isn't reality. A simulation of love is not love. A simulation of compassion is not compassion. It's just a computer game. And it can be very convincing, but it's not the same thing. So when we think about this future, we have to think about all the stuff that's happening around us. And there are three revolutions and the K makers I mentioned earlier, but these are the three revolutions that will change everything in the next decade. The digital revolution, of course, the sustainability revolution. 100x of the digital revolution. The biggest change in society in the last 100 years, going from fossil fuel extraction to creation, circularity, sustainability, green energy. That's a shift, roughly 2 or 3% of global GDP for the next 20 years. And the oil companies know this. They know, of course, they're doing their best to skid all of the results of voting or whatever can be done at COP28. But they know this is coming and they're going for as much money as they can make the next two or three years. The last one, of course, is a hard one. This is the purpose. Revolution. Why are we doing this? What's the goal now? This is being driven by the millennials, the Gen Y people, roughly between 25 and 40, maybe even 45. They are asking other questions. Our question, my question, my generation was better job, better money, more ownership. I mean, my generation created 50% of all CO2 that is in today's atmosphere, was generated in the last 25 years. That's our accomplishment. Now we have people saying, especially post Covid, we want a world that's a little bit more equal. Not just profit and growth, but people purpose. And these people, when they're graduating from university, they will not work for a company that supplies services to oil and gas. In fact, many universities have blocked recruiting from oil and gas companies. And you see many initiatives in this direction. This is a different world. And this is only 20, 37 years from now, if you're not sustainable, everybody will hate you. And the oil and gas industry is turning into a criminal minefield in the true sense of crime. I'm not making this up. We get into climate emergency. Emergency means emergency regulation like Covid, are you ready for a frequent flyer tax? To eat meat seems impossible now. If you suggest that now, you'd be off the table, Especially in Italy or in Switzerland, where I live. But it's emergency situations, extinction, rebellion. UK times 100, many people that live in London. My friends are saying, I'm thinking I should sell my SUV because it's dangerous to leave it on the street. So we are rebooting our world in technology. There's four things happening in information technology. Biotechnology that allows us to create synthetic products like airplane fuel, sustainable airplane fuel, cultured meat, vertical farming, 3D printing for each of us on this 45,000 kg of concrete that we have already created and used 45,000 kilos per person. And now we can solve this. It's more expensive, but we're going to figure this out very soon. In 10, 15 years, many cities around the world will have a high rise. And in the building, we will have a cultured meat department that creates protein based on laboratory fermentation. Sounds far out, but it's working. And vertical farming of organic vegetables in one building that feeds 100,000 people. No transportation of the food straight from there out again. Sounds like science fiction, but it's coming closer. And now AI is basically, we can say we're moving into this digital tunnel and that is going to create many, many, many new jobs in the US. After a lot of back and forth on the climate issue, they have finally gotten with the program. And Biden has announced a climate corps, like a Peace Corp 20,000 people will be hired to work on the climate. I think in every country in Europe, we should hire climate corps, employ people who would do that work anyway, because that's what they want to do, put them out in the street to fix those issues, to find the issues. And the key question really is ultimately, as we're going to that future, and how far do we want to take this digital convergence? And this is the question we must answer. Because today we can't upload our brain to the Internet, at least not most people. But this is becoming possible. Should it be considered normal? Should I be able to swap my legs against better legs, artificial legs? And Kevin Kelly, who is a famous futurist in America, one of my futurist colleagues who started Wired magazine, he says we should be optimistic. Not because we have less problems, because we have more capacity to solve them. And that is just so true. AI is one of those capacities. Climate change, for example, will be heavily impacted by what we can do with AI. Lots and lots of benefits here. We have to think of this from a different viewpoint. Our increased capacity. Here's the curve of climate change, innovation, climate technology. And you can see this is all happening in the next 10 years. So agricultural software, battery storage, climate Fintech or smart grid fuel cells, smart meters, huge list. We're going to share the slides later so you can review that again. Tens of millions of new jobs here and lots of money, all of the money used to go into oil and gas is moving here. We need to have different skills for this too. I mean, to work in climate change, you have to beef up your understanding of what it is. The other problem is currently the oil and gas companies put out roughly $2.5 billion of profit every single day. And that is buying a lot of votes and a lot of speeches. That allows them to send 300 people to all the COP events to make sure that ending fossil fuel does not show up on the agenda. I think some of that money from climate change, $5.6 trillion of fossil fuel subsidies every year, should go to the UN to address this problem. We should shift that money to actually make a difference, not to create more differences, worse differences. Because you see, the agenda is really quite clear here. And the problem is also quite clear. 3 degree global warming. This is the map. And right now we're headed for 2 degrees. The map looks slightly better, but still enough to cause nightmares. If you live here, some of you are from Africa. This is a red zone and we can grow grapes in Stockholm. We're already doing that bulb temperature over 32 degrees and high humidity means you die. This is a scenario that we can envision if that happens not too far and much quicker than we thought. And yet we're looking at this at the same time. All the climate meetings, all the agreements curve goes up. Just goes up and never goes down. Here I was just in Brazil four weeks ago. The Lula government, green left wing government, is about to give a license to oil companies to drill off the coast of Brazil to get $200 billion per year. Brazil wants to be number three, to take out more oil. So here's the problem. The reality is that, okay, we have all the tools, we can fix this, but will we have the telos, the Greek word for wisdom? The goal, the collaboration with AI? It's the same thing. That's why I'm bringing this up. We have all the tools, but will we have the wisdom to use them? Right? Looking at this chart here, this shows that we're going to be much more productive using artificial intelligence. For example, paralegals for X as productive office and admin support. Because these are power tools. It's like before email. Email made life a little bit easier for some time at least. Now we have systems that write emails for us or they can translate when I'm speaking. Makes it much faster. So I can work four times as fast. But what are we going to do with this four times as fast? We're going to say that's great, Gert works four times as fast. Everybody else can leave. That will be a bad result because I'm still working the whole time, I'm not getting any of the money. So you see the curve of productivity in the US Productivity goes up and employees salaries go down. That's because we're not using the right policy approach here to build a good future. We don't want an arms race of artificial intelligence. Whether it's capital against labor, that's also an arms race in a way, or just in the military. This is not the fault of AI or AI researchers. That's just what happens with technology, with good technology, exponential technology. The hardest thing to change, and this will be our main mission with artificial intelligence, is our economic logic. It's much harder to predict the end of the world than the end of capitalism. And the whole definition of capitalism or any ism is now kind of moot. Because what exactly is happening here? What exactly is our objective? I think we're moving to a world that is going to be powered by technology, allowing us to create this different emphasis, people, planet, purpose, Prosperity, not just profit. And that has been discussed for 50 years. People plan a profit, but now we have the tools. So what are we doing with those tools and how can we make this happen? How can we change the economic logic to create the good future? The Hollywood strike of the writers ended two weeks ago was a great example. The Hollywood studios said, many of my friends work there. They said, okay, we have all the scripts ever written. We have the ones that we didn't accept, they're also here. We feed them into something like ChatGPT and then we say, can you write a show like Dallas based on all the scripts that we've seen so that the machine can be the writer, which means the studio owns the script and then the writers come in and they get paid to clean the script. That was the idea. Of course that doesn't really work because it creates the same thing again, but even worse than the original. So that didn't work. And of course, ultimately they resolved the strike and the scriptwriters won in a big way. Much more than they ever hoped, I think to have. Christine Lagarde says automation is good for growth and bad for equality. If we're going to have more automation, then we have to figure out how it's good for equality. We can't just say it's automation, it saves money, it's good. That creates conflict, creates other things that we don't want. So how are we going to figure this out, this transition from this one minded Milton Friedman agenda? Now let's talk about AI and what that all means in that context. Basically, AI can be defined as computer systems that turn data and information into knowledge. That's from Damon Sasabes, the CEO of DeepMind, now owned by Google, one of the leading scientists. Think about this for a second. This is not the altered definition of machines do what humans used to do. It's information and data turned into knowledge. In other words, now it's up to the knowledge workers who get impacted here. Because if a machine can have knowledge, what do we do? Isn't that what you think makes you? I mean, that's what I used to think. I have knowledge, I have experience. What's good? Nobody else can have that if a machine can have that. I mean, my typical example is I used to write research reports about the future and sometimes we sold them for €5,000 a piece. And now you can go to Google Trends and you can go to IBM, Watson and you get the future of Switzerland instantly written for you, including a slideshow, a made up video, everything that used to be my job wasn't quite as good probably, but pretty close. And now the question is if a computer can do this. That's really what a computer does is the cogwheels. It does not see anything else around this because it doesn't know what it is. So when a computer says Gert is angry, it knows that from my face, but it doesn't know what angry is. It doesn't exist. And therefore if you don't exist, you can't have emotions. So it's basically just analyzing all of that. And that's quite useful. But if we look at this, I always say knowledge without wisdom is like water in the sand. It's an old saying from Malawi. So basically computers can have knowledge, but it goes into the sand very quickly without any wisdom, which is what we should be providing. And that's ultimately, I think the direction that we're going with this. Basically AI is a general purpose technology, General purpose computers, electricity, the printing press. That's important to realize. This is not a minor thing like the Metaverse or Bitcoin or blockchain, which are not minor things. But interesting nodded like this. This has been going on for 50 years that is basically impacting every single part of our lives and economic value chains and pretty much everything around it. And this deserves a holistic approach because if we don't take a holistic approach, we end up saying, okay, that's great, we can cut out all these workers and we can hire less people and we can teach our kids with robots and so on and so on, we save lots of money. But that's a very simple cut down approach to economic realities. So we have to take a wider view and figure out what exactly are we going to do here and which way are we going. I made a film two years ago called the Good Future. It was filmed in Lanzarote and it's quite popular on the Internet. But I couldn't figure out how to get other people that didn't speak English to really watch it. And now there's a software called rasc and they contacted me and I tried it out. And so now I can dub my films in 20 languages using AI, using my own voice. Here's Italian, You get the point. It's rather hectic there. It works perfectly in French and in German and in Spanish. Italian, it seems to be more difficult. I don't know why, but it's still entertaining enough. If you don't speak any English, you can still watch this and kind of get it. So it's useful. It is not dubbing like a real person would dub it, which would cost €20,000. This is what AI does. It allows us to leap. And then we say, no, that's not. That's garbage, right? Or it's great or it's good enough. But I tell you one thing, the difference between good enough and really good is a canyon, a United nations translator translating what I say right now, that's the top of the line. AI doing that is suitable. Sometimes it's funny. Is it going to get perfect? Highly unlikely that it's going to be as perfect as a human possible, yes, but it takes a lot like background noise and all these things that machines don't like. I also built a bot for my own website called TheGurge AI. I put up all of my books, 1000 videos and 2000 blog posts. And the AI is searching inside all of those for answers. So you can go to Thegerd AI, you can ask a question and it will give you an answer. Very soon, starting next week, I will actually speak to you in my voice because I want to see what it does. And it's really interesting because sometimes, most of the time you get pretty interesting answers. If you don't know by working, go in and say, what does Gerd say about future of China? Or something? And you'll get an answer just like you were talking to me. But if you go into my material, my blog post, I write about other people's work. And sometimes I write about XYZ professor who has written a research paper. If he asks the machine, has Gert been awarded a Ph.D. he says, yeah, you're 14. It just doesn't know the difference between me speaking about other people and speaking about myself. That's what we have today. As was said earlier, we have to keep the human in the loop. This is the crucial part. Even if it's more expensive, more clumsy and takes longer, because that's the only way right now that we can deal with this Google Maps problem. We love it, we use it, but we say, no, no, no, this can't be true. The shortest path isn't over this bridge. Great example. Now we have artificial intelligence assistants that can basically become your own personal digital assistant. This is the same guy, Mustafa Sulaiman, who started PI AI, which I showed you earlier. See what he says about the future.

Everybody is going to have an intelligent assistant, a personal intelligence that knows you, that is super smart, that understands your personal history and can actually hold state. It can preserve things in its working memory. So it will be able to Reason over your day, help you prioritize your time, help you invent, be much more creative. It'll be a research assistant, but it'll also be a coach and a companion.

Okay, you wrote a book about this. A coach and a companion. I mean, that's completely different things here, right? If we're going to have a personal digital Assistant, imagine that 5 billion people connected to the Internet, not just on your mobile phone, but in your buttonhole or your wristwatch. And this becomes like a person, as we've seen in the movies, like Black Mirror and others. Is that a good idea? And if so, who finds out what is real in here and how does it actually create a different world, a different narrative? So we have to look at this topic of humans versus or with machines. And this is one of the key slides. As we're moving into that future, it's quite clear that machines are starting to pick up the lower part of this equation. So digital intelligence is very good with data and information and some kind of knowledge. That's why I put only the half line here. Give an example. If your son comes home after school and he's 13 years old, you look at him and he says, I've gotten bad grades. I missed the bus. And so that's called data. But if he doesn't say anything and just smile stupidly, then you know he's fallen in love for the first time in his life. That's called understanding. It has nothing to do with data, because he didn't say anything. Because you've been there, because you're up here, and this is our future. Machines will learn all that stuff about everything, every time, about anyone. It can be very scary by itself. But basically, we're going to move into a world of more biological intelligence. Following the analysis of various psychologists who talk about human intelligence being eight or ten different things. Emotional, kinesthetic, social intelligence. That is, for the time being, impossible for machines because emotional intelligence is not even defined in humans. Allegedly, women have more emotional intelligence. I'm very hopeful for this crowd here today. There's so many women. That's great. And if we're looking at education and training, we have to focus on this. So not just beat the machines by being smarter. I'm a better programmer than the machine. That is just not going to work because the machine can program, can do anything that doesn't require this. So we should not be creating jobs or positions or trainings that focus on this. I think it's a fundamental change in how we think about the future of education. And learning. We used to think when I went to school, we download information for later, keep it stored somewhere, and then when the right time comes, I pick out the case study and boom, here we are. Everything we've known about the world so far is under fire. We have to learn how to learn. Alvin Toffler said, unlearn, relearn and learn again. And that is going to be quite a challenge to change this, because in this world, with those two intelligences where we have digital intelligence that's in the cloud, we can have unlimited supply of this. That's the good part. You want to learn how to fix your motorcycle. That's all on YouTube. All these things are there. And of course, on the other side, biological intelligence is stuff that is not so easily accomplished. Emotions, creativity, imagination, ethics, empathy. We don't go to school for empathy. I suppose you could go to some schools for empathy. But this is something that we learn in life and this is about traits, character traits. And this is what we have to teach and this is what we have to think about, what we teach ourselves and our kids. The roboticist Moravec said, whatever is very simple for a human is very hard for a computer and vice versa. And that is the reality of AI. And furthermore, that is also because why we shouldn't build an AI that can be like us, because then it could do both. That's not going to end well. I mean, think about an AI that has an IQ of a billion. An iq, logically of a billion would anticipate every move that every human could ever make in the next 5 million years already mapped out. So it doesn't strike me as a very good reality. We have to think about this by being more skeptical of reductionism. And a lot of tech companies are going down this road by essentially saying, if you buy this, your life is solved. Connect more, makes you more happy. That's just not true. If you're not connected, maybe you're unhappy, but if you're over connected, you're also unhappy. And if I buy the Apple Vision Pro, I can work faster. But am I still human? And why does it matter? It matters to me, but some people would argue it doesn't. This is the worldview that we are. Organisms are algorithms. That's the worldview. I think that is just completely off the wall as we're looking at this kind of idea of what we are. Logic alone is not enough. The idea of only having logic leads to a terrible world. We've seen some tyrants and a lot of logic ruin the world. For everybody. So we have to look a little bit further. Algorithms know the value of everything, but the feeling of nothing. What is life for humans? Experiences, relationships, engagement. You wouldn't answer the question saying, my life is about data. These are important questions. So as we go into this, we have to think about policy. We have a significant problem. As mentioned earlier in other speeches here, we're investing trillions in developing technology. But what are we investing in? Human alignment. Making sure the consequences are feasible. Making sure that the consequences of technology are captured. That's called bad alignment. Here's what bad alignment looks like. Imagine a machine that you would tell to do something and it would do this. 90% of the time. It would be unusable. And this is what we have with AI. We have to check it. We have to have standards and maybe some regulation. Maybe. Think about how that works. Misalignment. Disaster. Space Odyssey. Open the pod bay doors, Hal. I'm afraid I can't do that. Remember that scene? The AI takes over the ship because he wants to unplug the AI to do his own thing, and the AI says, I can't do that because the plan is different and you're no longer important. We. We don't want to get to that point by using AI. What happened in the last year that.

Made this so urgent for you?

So for a long time, I was working on making computer models that got more and more intelligent in an attempt to understand what went on in the brain. And very recently I realized that the computer models we're now making may actually be a better form of intelligence than.

What'S going on biologically.

The idea that they might overtake us quite soon suddenly became much more plausible. For a long time, I'd said that would be 30 to 50 years, but now I think it may.

This is Geoffrey Hinton, the godfather of AI, who started deep learning and machine learning. There are many people who don't agree with what he says. But here's the bottom line. When we build a machine that can do all these things, how do we know what actually goes on inside? And we already don't know that with ChatGPT. We don't know what goes on inside. So if we let that machine make decisions about the stock market, it could lead to a stock market crash. Because basically we would let the machine influence us in our buying decisions, not give it access to the stock market. So that's the misalignment that I'm worried about. What we have to do is we have to think about this like this. On this Kind of sliding scale. I think we're fine with assisted intelligence. And this is primarily what you guys are going to do in your daily job. It's basically you work on steroids, making it four or five times as efficient. That's probably a good thing in most cases. Some kind of automation or augmented intelligence with virtual reality. But autonomous intelligence, I think that will be taken a little bit far. This is why I've been talking about this for 10 years. I know the UN, a similar concept a couple of weeks ago. I call it the International Artificial Intelligence Agency, the iaia. The agency that does what the nuclear agency does and basically figures out how to create rules. Not that we need this for every minor thing, but we need it for the big thing, which is controlling general intelligence. And this was just announced a few days ago. Multistakeholder advisory body, I think will be announced who's actually in it very soon. I think this is our problem. The first part of this is really beneficial. It makes us faster and better. IA means intelligent assistant. The third part, we probably don't really want to do that. A super intelligent machine can make up a pathogen, for example, a biological weapon, just by feeding all of the information of chemistry into it and coming up with the perfect mix just as one crazy scenario. We need to figure out how we can stop that from happening. I think we need to pursue competence in AI, not consciousness. That's why I'm worried about companies like OpenAI. Because their goal is to create this general intelligence. Why is that their goal? Because they want to be good to humanity? No, because the biggest business opportunity ever is to rebuild humans, Literally to build digital humans. We need to think about this. When you think about human resources and your own career, technology can do great things, but it does not want to do great things. It doesn't want anything. Says the CEO of Apple. If we want technology to do great things and good things, we have to make it do great things and good things. We have to create rules and regulations and also social contracts around it to make it work. Technology by itself has no ethics. Some technology companies want to have ethics, or may have ethics. That's not the same thing. But it's difficult. You're looking at $2 billion of revenues versus 500 million with a little bit less ethics. Maybe you make another decision. We have to think about this. The guy who runs AI at Facebook, he says we're going to have a future when everyone uses some kind of digital assistant. And we're not talking about 50 years, we're talking about two or three years here, how people are going to make up their mind, what is the narrative? Who controls the story? What's the truth? How do we vote after we talk to our assistant? That is a concept that we really have to think about. And the problem is that what we have right now with AI is that AI is basically out there to kill truth, because truth does not matter. For a machine that has a binary approach, yes, no, yes, no, yes, no. It just finds the most obvious fit for the answer. True or not, it fits. That's why two days ago I stopped using Twitter. I've been a Twitter user for 10 years. Because what I see on Twitter is exactly this dishes up things that fit the scenario, true or not, and it allows the creation of bots that serve up stories that are completely made up. And my stories that are real are going to feed the engine to serve stories that aren't real. So I decided I should leave that behind. The end of routine this is what's happening in work. I mean, if you're in HR or if you are doing work related things, this is clearly what's happening. As we're moving to the world where you can see this, all of these things are nuts and bolts. Things that AI will help improve operational efficiency, customer experience, employee productivity. All of these things called better software. Basically, they're not rocket science, they're not miracles in our chatgpt, they're not, they're not transcendent and nothing like that. They're basically just better software. That's why I call it IA Intelligent Assistance. Doc Engelberg, the founder of AI, he said great technology should aim to not replace humans, but to amplify them. And to do this, we need to also put a stop to trying to replace them because they're taking away the power from everything else that we could be doing. If a pharma company can run medical experiments about new medications in the cloud, it can cut down development time from five years to one year and roll out vaccines and everything else much quicker. There's a benefit that's augmenting human things. That's the powerful thing about this. What's happening is that machines can do the routine if we let them, and they can help us be quicker to respond. And I think part of that is this feeling that people have about routines, is that we become useless humans. But if we take routines and the machine does it for us, what does it mean for our work? It's not the end of work, it's the end of the routine. If Your job is 95% routine like a call center, you are in trouble because there won't be much left. 23 million people around the world. But then again, it's not that simple to say that every call center will be out of work because there are many angles to this job. Driving a car is not routine, it turns out, because we don't have self driving cars except for San Francisco, we have to have it there. But anyway, it shows you that it's not so simple to drive a car. And maybe it's not so simple to be a good call center agent, as it appears. For example, really understanding people involves more than language. Maybe we could replace the pilot, but why should we? We're talking about one or two people here. Does it really make a difference? And what else does a pilot do? Many things that have nothing to do with flying the airplane. The machine couldn't do that. The machine is a bot. But then you see the chart like this showing that lots of our jobs, and I'm talking about our jobs here, have potential for automation and potential for augmentation both. And we don't know how that will play out. It will play out in such a way where all of a sudden I have something that's being automated and makes a little gap in my work schedule. And then I decide, hey, I have some more time. I'm going to study XYZ so I can add more value or I'm going to go play golf. That's another option, not to work more. But like in France, the idea of working three or four days a week, six hours per day, because of this, that would mean that we get paid the same. That's of course the big discussion. In other words, the productivity, the value is there, but the policy isn't. Policy says we keep the money, you work more. That is not a good idea. It will not take us to the final destination. Let's talk about what that means for education first. We are surrounded by technology. I call this the Neoluvian man, like the Vitruvian man, not far from here, a long time ago, Leonardo da Vinci. We are surrounded by technology and this isn't going to stop. There's no way that we can say we don't want this, we don't want AI, we don't want smart technology, we don't want the Internet of things. We don't want. Yeah, we can do that, but it's hard to get away from that. Not really an option for most people. What we have to think about is how do we combine this with that. That makes Us, human. I call it the algorithm. So creativity, imagination, ethics, consciousness. So in some cases, I think we need what's been called the rewilding, which is to say, let's go back to what actually makes us human. At the same time, offline is a new luxury. In the old days, we used to escape reality and go on the Internet. And now we escape the Internet and go to reality. The things that only humans can do that I mentioned earlier in the pyramid. That's the future of work and training. All the things that we didn't want 15 years ago. Inefficiency, creativity, asking questions. Buckminster Fuller, a famous futurist, once said, we go to school to be de genius, to have our genius removed so that we can work safely. That won't work here, because work safely is what computers do. They're very good at this. So as we look in this direction, we can safely say, I mentioned earlier, technology has always created new jobs, especially with professionals. So the more innovation we had, new jobs came. Is that true this time? That remains to be seen. But I think if we're looking at the scenario, we have to get away from the thinking that this is the ticket. This is one of the tickets. Of course, it's great if you're an engineer, if you can program, but just being able to program, I could speak to my wristwatch and have it make an app. We have to bring back the humanities. I call it hacky, Humanity, ethics, creativity, imagination. Very, very difficult for computers to do this. Is it difficult for a computer to program? No. Increasingly it gets easier and easier. These are the logical jobs, things that we have to think about. And it's quite clear the jobs are going away. The sort of more mechanical jobs. Financial, insurance, public administration. These are jobs that can be commoditized by smart machines, smart software, as are some of our own jobs. The new jobs on the other side, healthcare, scientific, technical, communications. Also many of your jobs increasing. Why is that? Because it's very hard for a machine to do that. In a nutshell, if you work like a robot, a robot will take your job. Hey, that's not surprising, okay? I think most of us already realize this. If you learn like a robot, you'll never have a job. Don't let your kids learn like a robot. Tell them, do the opposite of the robot. They'll be safe. Ask questions. Picasso once said, humans are for questions, machines are for answers. So let's wrap up and take some questions. First is efficiency. Obsession to have a good work situation does not depend on efficiency anymore. As much as it depends on new things. Human agency, consciousness, imagination, intuition, empowerment, creativity, courage. That's also because of COVID We found out that actually these are the things we need to be resilient. The whole idea of KPI, I call it khi, the Key Human Index. That is what we're looking for. What are you looking for in employees? To become a true human. That adds value to the organization, not that performs. Robots perform right. Go beyond efficiency, resilience, agility, all of those things. And let's stop looking at the past as we're driving warp speed into the future. Europe is very good at this. I lived in America for 17 years and Americans are obsessed with this. It's all about the future. I live in Zurich. I feel like I'm in a giant museum. We have to change this. Let's look at the future. 45 minutes every single day. I'm not talking about watching films, I'm talking about reading, doing the research, talking to people, understanding what's happening. Because then you can formulate answers. You can have what's called foresight, not prediction. Prediction is much more difficult, probably impossible, but foresight, understanding what's coming next. I call this the future mindset. You would not believe how many people I talk to that just don't pay attention. It's unbelievable how people run big companies that don't pay attention. Not to their employees, not to their wives, not to their kids, to nobody. Because they're looking at the path towards the shareholder return. So we have to start paying attention to what's happening. This was used in World War I and 2 to hear the planes coming. And some of those were like 5 meter tall. So you can hear them. Amplify our ability. That's what we need. We need to amplify our ability to hear the future. Four things about this, and then we'll take some questions. First, observe, understand, imagine, and then do the future in that order. Many people don't want to go through the observing part because they think you can shortcut it. But I find the more I use AI, I find that I care more about my own research because it's kind of like cooking. I go to a great restaurant. I love what they're doing and I trust them. But my research, I can't rely on somebody else having done that for me. I need to understand all of it. It's not organic to just read something and say, oh, that's it done. It doesn't work. So we have to pay attention to these things. I propose that we do right here in Turin, the Good Future Congress, an event where we decide what we're going to do to build the good future and how we're going to do it. I mean, of course, this is the agenda of the UN in general. Sustainable Development Goals are all inside of these things I talked about. Clearly, very much related. So the Good future is about two things. The telos, the wisdom, the understanding, the purpose, and using technology to get it done. And that's where AI belongs. It belongs in this pot. And it's giving us a superpower that we need to use wisely. And that's why it's so important that we create this handshake in a good way, that we think and we ask questions, coordinate what it can and what it can do, and that we know when it's there and when it's not there. I want to quote Buckminster Fuller. We are to be architects of the future, not as victims. And this is so important because right now, a lot of people are feeling like the future is just happening. There's nothing they can do. Technology, clearly, the three waves. Digital, sustainable, and purpose. The human revolution. Holistic, circular, human. That has to be the goal. And that, of course, is inside of many things that are being done here. So that paradigm is coming. Sometimes it looks like the opposite is happening. It may get worse before it gets better, but that is the destination. The future is better than we think. Thank you. Thank you.