 to be here. You get my slides up, you know, great. Alright, thank you. So I've been to Kerala a few times. Last time, several years ago, I had a pressure on food and diabetic diet just down the street and I lost my weight back then, but I got it back as you could see. But it's a great pressure to be back in India. As you know, the future is not invented in Europe. The future is invented in the countries that are going to be the next big countries, not the countries that they have been called developing countries, but primarily India, Brazil, China, countries like this. And it's very interesting to see what we're seeing now. I said technology is now coming, boy, let's give them a little slide here, which has started at the beginning. Technology is now becoming a driving factor of society. If you're looking around, you can clearly see a couple of very interesting things like Google, for example. I do quite a bit of work with Google. Google is now becoming an artificial intelligence company. That's a scary thought also. Because Google knows more about you than your husband or your wife, because of all the eight years that you've put in all information to Google. And the mobile OS, right? Google has very deep information. Just imagine in five years, do you really think you're going to steal at a machine like this and type in best sushi in Trivandrum? The machine will already know who you are, what you want to eat, what your friends have eaten, where they're eating, and you just speak. I was in the Intel Labs last year, I was sitting on the couch, watching the next generation of television. And you sit in the couch and you say, I want to see House of Cards Episode 3, where he kills the guy in the garage with a car, and will play instantly. No searching, no typing, speaking, gesturing, thinking. Now, that sounds really good. Of course, there are many issues here. As you know, there's a lot of really hard things about human intelligence. When you talk about artificial intelligence, AI, there's so many confusions about this. Really, what AI is abulating humans in a very basic sense. And guess what? It is extremely difficult to emulate people. Language. It's been worked on a very, very difficult time. Very difficult. Vision. I mean, the computer can now understand that the picture on the internet is a catch. But it won't know if it's a dead catch for a live catch, right? Unless it's obviously dead. Motion. Computers can walk. It's very difficult to walk. Robots can walk, but it's extremely difficult to walk correctly. So there's many things that we're seeing here that are just now being worked on, and we're now at the point where these things are coming possible. They're no longer science fiction. So my work is a feature based on, I'll give you a bit of background here. And by the way, I currently live in Switzerland, but I spent 17 years in the U.S. as an internet entrepreneur and also as a musician and producer. So if I speak very excitedly and use the word awesome, that's my American part, right? If I speak very gently and slowly and don't take any risk, that's my Swiss part. But I was born in Germany, so if I put fear into you, that's my German part. And that helped me a bit of fear today, right? So basically as a futurist I do this, right? I don't do predictions. All of you could do the very same thing that I'm doing. You just have to inverse the work that you're doing on today to focus on tomorrow. All of you know what's going to happen in five years, not 15. All of you have a good idea what's going to happen in five years, but we're very busy working today. So I reversed the time. My clients spent all the time working on this quarter and this year, and I spent all my time thinking about what could happen in five years. So it's really an observation. It's a great saying in China where people say if you want to know about the future, ask your children. And so I would give that advice to you because the children will tell you what the future is, right? My company, the Futures Agency based in Queensland also in San Francisco, our motto is it wasn't raining before Noah built the art. You might know Noah from the Bible, the guy that built the art to save mankind. It's very important to find out that what is happening is a lot of companies are under significant pressure to reinvent. And you know the reinvention started in one place that's very dear to me, the music business. Remember Napster? Late 90s, 75 million software downloads in four weeks for free music. And so first music was disrupted and then movies and films and then the publishing business, news, magazine, print, and guess what's number four in that industry? Financial. Financial is becoming like we get. We would see the same disruption and the financial business that advice back came money, transactions that we see in music. So we were tremendous changes and it's very important for people to anticipate what those changes are. You know that the music business was famous. There's only four people who ran the music business. It was essentially a cart cattle. So Sony, Warner, EMI, and Universal. And when the internet came around the late 90s, the record label, I was working with the record label, they said we don't like the fact that the consumer can do what they want. You know, download for free and stream and share music. We will try to stop this. The music industry's suit went to court for 259,000 people in 10 years. And only one of them went to trial. And they lost 72% revenues and 10 years. Imagine if they were to spend the energy on thinking about the solution for what people want rather than stopping people from what they could do. So I would advise you if you're in the business of financial or any other technology, you have to take a look at what people want and stop it. What we're seeing right now is, of course, computing, social, mobile, intelligent assistance. We're moving into a world that sometimes may look like a time fiction. On that note, please, you people are always, most of you, I'm sure, you are really into motion pictures. I am. I watch all of those movies, you know, from Oblivion to Transcendence to Ex Machina. When we talk about artificial intelligence, ambulating humans, forget Hollywood, please. This is entertainment. It's meant to be inside fear and, you know, it's meant to entertain, right? What we see in those movies is not what is going to be relevant for us. It's always the worst case scenario. Role was taken over the world. Yes, there is a potential for that, yes. Now, we could also have quite hitting the earth, right? So, please, forget about Hollywood. Let's go back to what we've seen in our real lives, okay? What we see in our real life is this, and you can see this in India in all the big cities, the convergence of man and machine. The convergence of what we can do with people and what technology does for us. Our technology is extremely empowering, like social media. We can go and use TripAdvisor and say bad things about the restaurant. We can go to our college and bring our professor so that the other people can... We can make a comment on Facebook. We can research things and prepare practices, right? It's empowerment. At the same time, of course, now we have artificial intelligence already in, like, 50-plus. When you use Google Maps, that is artificial intelligence. Google Maps says, oh, I know Manish, right? I know where he's been, or he likes, who he's connected to, where he is, and I make a match and I send him the right messages, right? That's kind of intelligent kind of a very simple level. But you know what happens when we need in person? When any of us need in person later, it takes an average of 3.5 seconds between us to exchange so much information, and that information is not outspoken, like unspoken, that the computer will take three months, any computer, to deliver that much information to each other. Because there's many things that we don't say that we mean, and there's many things that we don't mean when we say them. And so it's very important to remember that what we're trying to do is not to emulate people as we are people, that is not going to be possible for quite some time, at least I hope. But to use technology as a utility, not as a religion, there is a difference. So when we talk about artificial intelligence and talking about utility, we talk about one of the most powerful tools given to us in the next, well, it's already here, but in the next 5 to 10 years, we'll revolutionize pretty much every single thing. And we should keep in mind, just like when you use TripAdvisor, TripAdvisor is widely used in India, right? Well, you know what it is, right? Rating restaurants in hotels. When you use TripAdvisor, sometimes when you go to a hotel or a restaurant, it's 100% correct, right? Not very often. Sometimes 100% wrong. So TripAdvisor is a data partner. If you were to live your life based on TripAdvisor, you'd be in deep trouble. It's good, but you would never live your life based on TripAdvisor. Would you marry the right woman or man based on the DNA analysis? Which is now possible, right? We shouldn't make the mistake of saying that the data that we get seems like the truth, right? It's just one piece of it. Let's keep that in mind when we talk about artificial intelligence. So Ray Kurzweiler, who is probably the most famous futurist, and I would urge you to read his book called The Singularity is There. I'll explain later what that means. Ray Kurzweiler says, in Singularity 100, basically the future holds for man and machine is convergent of man and machine. And of course, he's very, very extreme representative of futurism. He says that we will connect our brains to the cloud. Basically connect our brain directly to the internet. That is actually his proposal. So here's something very important to understand when we talk about Ray Kurzweiler's Singularity and the future of artificial intelligence. And it's called the Morovedge Paradox and you can look it up on Wikipedia. But it's a very simple idea, OK? What is extremely easy for the computer is very hard for people. And what is very easy for computers is very hard for people. OK, again, sorry. What is easy for computers is hard for people. What is easy for people is hard for computers. So a computer can be pretty much anyone in chess. Has been for many, pretty much anyone. But a computer cannot talk to a two-year-old child, which is not very intelligent in the sense of, you know, playing chess. And so that is our challenge. When you're engineering software, you're always going to be up against this one point. What we as humans do is very hard to copy. What machines do is actually very good for us to have. For example, a doctor having IBM Watson, you know, making rounds in the hospital. The doctor can have access to 100,000 cases of the same cancer through the machine. Very powerful. So that's important to remind that paradigm. Now also, we're living in an exponential world. There's a great book that just came out by a friend of mine called The Exponential Organization. It's very American. It's very, like, you know, pro-technology. But you should read it. And the guy's called Louis Vermeer. He's part of Singularity. Now this curve is probably the most important curve in your life. You know what it means? It means that we're no longer at Moore's Law in the beginning of this curve. We're no longer over here on the left, you know, when you're on the left in the mid-19th on the Internet. If you double 0.0.0.1, you still have nothing. But now we're doubling 1, 2, 4, 8. And we're now at 4. The next point after 4, Moore's Law, right? The next point after 4 is 8. So it's a mind-boggling speed. Now we're at the take-off point of exponential technology. That's not just artificial intelligence. Bioengineering, nanotechnology, geoengineering, solar power, 3-D printing. I mean, if you put all these things together, you think you've come straight out of Harry Potter, or some movie like that. So you're very lucky to be at this point. There's very many business opportunities. But you also have to keep an eye on the side effects. Most of the development in the next 5 to 10 years will include some computer-ambulating improvements. I don't know if you've tried the latest version of Siri, but it's just coming out. Even Apple has just created an update that's coming out. You may have seen on the WWDC video, the introduction. Basically, this system will be capable of speaking to you like a person. You can speak to Siri and say, last year or last time I was in Kerala, showing all the pictures of me and my wife on the beach. It was like a person. That's all happening coming within me. Imagine what that could do for personal services. You're saying, I want to invest in real estate in Sydney, Australia. Can you find the places that are closest to the beach not yet populated with rich individuals that have an average weather temperature of 26 degrees and where there's lots of Indians and maybe some Germans. Go find this. 14 seconds later, boom, right there. Huge report. Send to your brain with a wireless interface. Maybe not. This is the company in that Google has founded for Facebook called Vicarious. They spent $150 million. The company is funded with $550 million. Look what they have like. We're building software that thinks and birds like a human. I mean, only Americans would say that, right? I mean, if I would say that, I would say we're building software that copies what it means to be human or tries to parallel that. Can you build software that thinks and birds like a human? That is an extremely tall border, right? That is like saying you want to send to the machine, conscious machine. The Google self-driving car, which I had the pleasure to drive in a few years ago, the self-driving car by Google is an extremely complicated and smart machine. But this car, when it goes down the road, pretty much anywhere now in California, you have a double yellow line. You have a frog on the road. The car will see the frog, and if you're lucky it will know it's a frog, right? But it will say it's a frog with a double yellow line. I cannot cross the line as forbidden. I cannot kill the frog as forbidden. I cannot endanger the frog by going over it very slowly as people probably die. So what does the car do? Shut down, right? Hand of the road. Cannot decide. Do you know what a human driver would do? One tenth of a thousandth of a second. We will make an instant decision what to do. Kill the frog, cross the line, do something. It's not important, right? There's a frog crying out loud, right? Yeah, you wouldn't keep the frog alive. You would kill the frog before you kill a schoolgirl, right? So that is actually quite difficult in the software. There's a great story about exponentiality that you may know is actually an Indian story. When the chess board was invented, the inventor came to the, it was a Muti Empire of 504 Price. He came to the court of the emperor and he said, here's the chess board, and the emperor loved the chess game. They made a bet. And the wise man said, well, if I win, I only want to be able to feed my family. You may know this story pretty well. But the long story short, when the wise man said, I just want you to put one corner prize on the first field and then double the second field and then double for the third field, right? Exponentially for each field. And the emperor said, oh, you're really a honest man. It's agreed. If you win, you get double as many rice balls for each field. In the midst of the game, he realized this. By the time he got to the middle of the game, he already hoped the guy was about to win all of the rice in India. All of it. At the end of the game, he would have owed the wise man one meter of rice covering the entire globe. Like 150 trillion points. And this is what exponential technology is. And guess what? We're here now, right? We're in the second part of the chess game. We're at the point where it's exponential before, right? There's not just little stuff like 1, 2, 4, 8, right? This is 150 trillion, 300 trillion. That's what's happening with science and technology. Humanity will change more in the next 20 years, like I said in the video, than in the previous 300. Half of these changes are driven by technology. The rest is driven by economics, of course, which is related, obviously. So in technology, it used to be, when I first got on the internet in 1995, the question was, can we do this? How much will it cost to build this, right? That was the question. You know what the question is today? The question is not whether we can do it. The answer is always yes, always. Can we ask them? Mind the asteroids? We can. I mean, it's hard, but we can. Can we set up solar systems around the Earth and transport the energy? No, we can, right? It looks like we can. Can we change the sea water into sweet water? Now, we found technologies. We can. This mind of ours has changed, right? Here's a great example of Uber, right? The taxi company. You guys know Uber, right? It's basically you have an app on-demand car like a taxi. It's a very big change. And Uber has announced, of course they're a American company, you know, speaking about how they were to dominate the world, but basically Uber has announced that they will have self-driving autonomous cars that they're going to invest in. Here's an astounding number, right? The Columbia University study says that 9,000 cars, automated cars, could replace all the taxicabs in New York City. And people would wait only 36 seconds and pay 50 cents per mile. Faster, cheaper, greener, more efficient, right? And they would lose 185,000 taxicabs. That's going to be a long time in India until that happens, right? But for obvious reasons, the traffic is slightly different there, but that's how exponential changes happen. So this book they're talking about, this is one of the graphs from the book, the exponential organization, they're talking about the sweet spot, this is going on right now. You're in the sweet spot right here, at the sweet spot of exponential is right at the place of disruption between linear progress and exponential progress. That's where software is at right now. We're now at the point where we can use that gap between the things that used to work linear. And my friend, my friend Frank Diana, who worked for Tata in Felte, came up with this really interesting list of things that are happening around us now. You probably know most of those. Autonomous vehicles, a shared economy, connected health care, smart cities, 3D printed. I mean, this list, you can download the slides later, by the way, there's lots of information here. So what's important to notice with all these topics, it's not just exponential, it is also combinatorial. So that means everything on this graph happens at the same time. The internet of things, big data, social media, cloud computing, normal crisis, cognitive system, robotics, they're all coming together. So if you're building software, you need to know about all the things around it. That is the challenge I think that we're no longer just about one thing. And artificial intelligence, of course, is a part of all of these changes. So now our software business is jumping from the little little bowl that we were in before to a much larger bowl. Mark Antrieson, the founder of Deathgate, and now a venture capitalism in the Bay Area, he said already four years ago software is eating the world. You know, that's very true. Everything that we used to know that's hardware is now software. Music, films, books, magazines, and people are still reading printed magazines in India. True. That's a whole lot. In the US and in Germany and Germany. So everything that used to be hardware is software. And everything that used to be software is now called an experience. Like Airbnb. Like Weber. Like Dropbox. So Marshall McLoone, who was a very smart futurist, he said it's the framework that changes with technology, not just the framework, not just the picture. So if you build the software for artificial intelligence with services for your client, you could not think of one tiny picture. You have to think of the frame. That's where the opportunity is. If you're very good at building pictures, that's great. But we're talking about an ecosystem here. When the banks are changing to a great digital and the millennials, the people who are about 30 now are going to buy those services. They're going to buy in an ecosystem. They're not just going to buy one tiny thing. So we have to build an ecosystem. We all must look at this. What is happening with computing? When I started using computers roughly 15 years ago, I was actually pretty late with this. You had to tie a button to the front of a box. Most of your kids will not grow up with the computer being a box. But where's the computer? It's in the air. It's everywhere. The computer's in the cloud. It's like water. You speak to it. I said, it's a scary concept, right? In cloud computing, basically, we don't need devices. We just connect in any which way we can. So what is happening here is we're going from typing, to speaking, to gesturing, to blinking with our great computer interface to thinking. You think that's science fiction? It's not. I'm not sure I would want to connect to my same thing, right? You have to be the judge of that. But the interface has changed so quickly. Microsoft mobile has seen that, right? The Oculus Rift. That's for geeks. In five years, as normal as a mobile phone. So in five years, if you're a financial advisor or insurance company or whatever, you wear augmented outfit. You can deal with so much data in one minute. It used to take you all day to do a knot on the computer. Remember the scene from Minority Report when it goes inside the data like this? It takes the data out like this, but over here and, you know, structures the data. If you're a doctor, you can reach inside of the data, right? That's how many happenings there's two times. So a great treat here from Alan Levy a few weeks ago. He says enterprise software used to be about making existing work more efficient. A lot of you are involved with that. That's good work. But the future is to transform the work itself. The future of artificial intelligence and software is not just more efficiency. That is a good thing. You can make money with efficiency. But is to transform the work India has many call centers? Any of them around here? You know what's going to happen with call centers, right? We're talking basically the language recognition that computers are able to do now. We give it five years and we can have 90% less people in call centers. I mean, there are so many companies I looked this up last night on the age when we see this. There's like a thousand people looking to kill call centers. Funded companies. You may say it's very hard, right? It's very hard to listen to people. But the stuff I've seen for call centers, and it's mind-boggling, there's been a test with a company I shall not name. But a hundred thousand people called this call center only four of them said it was a computer. Four for my hundred thousand. We're talking about very simple stuff like we looking at flyers, right? Not like therapies. Very simple things, right? So that's going to really change society. What are we going to do with all those people? Has a very big societal impact. So the other thing is when I was going back to the Hollywood motion pictures then when you're looking at the future of artificial intelligence you do not want to look at all the super human stuff, right? The things that are out of movies, right? Really what we're talking about is very important to remember we're talking about adding values to utility. That's the mission. The mission is not to copy humans or to make humans unneeded. You can try that. I think it's a stupid idea. I think in 50 years we can talk about that again. It may also be a very dangerous place then. But you know that people 98% of transactions are not based on logical decisions. 98% Car approaches this 100%. We use logic to make an argument that we should listen to what we want to do in the first place. Invest in these humans. We need utility. We don't need somebody's brain to replace ours. Again, you could try that. I think it's doomed for failure. That would be like saying every time I go out I go to where Trump and Pfizer tells me. Not good, I think. It's very useful. Very useful. But removing humanity from the process then based on what we're looking at here is intelligence, optimization. Use the technology to augment me. To make it quicker and faster and more efficient but not to remove my intelligence. To give me a kind of super intelligence. If you can give your clients super intelligence they will pay a trillion dollars for that, right? If you try to remove your clients from being part of the process they will kill you. What is the human purpose of being in business when you're not involved? And it's just a machine. So predictive analytics. That's useful. The system tells me that most likely I will need to have another 100,000 SKUs ready to go next week. That's very useful. But at the same time I always say, I think this is interesting but the system has kind of given me an inclination but I have another opinion. When you take this question, for example a really interesting question you know that all the major airlines around the world want to not have airplanes or pilots, right? But because they don't want strikes. They don't want people messing up and receiving threats. And you know it's pretty much proven that an airplane without a pilot would be safer. And it's pretty much proven that discussions are ongoing, right? But in most cases when the pilot has to take over from the automatic pilot he cannot respond so quickly. He fails to respond. That's been the cause of most accident. But having said that would you go into a flying robot to fly from your husband to Dubai? Without a person in the front? Without a person in the front? Would you enter such a flying robot? Well many of you say of course, and this for the argument is clear. The data says yes, but we're saying after. Maybe not. So there are, there is a difference there. And it's very important is that if you build this software and you argue technically things you need to think about the ecosystem. What is all the stuff that you're impacting? How do you change the food chain? If you're building intelligence for financial systems or insurance for government impact you're going to change the way that the world works for that matter. You're going to impact the whole food chain not just the piece of paper. I mean the fact that 1.3 billion people are using Google Maps is changing everything of the food chain. Where they go to which route they take how they can be monitored, surveillance. Take the example of Tesla. Tesla is the best example of a company that thinks about the ecosystem. You know Tesla the car company of course, right? Many of you probably have one, they're just kidding. Would kind of touch the drive I think in many places where you don't have a battery but it's an interesting car. But anyway Tesla is not a car company. They happen to hold some of their cars. They have a database they make technology now they make batteries tomorrow they will solve the energy problem distributing solar energy. You think of Tesla as a car company so do I, right? But they're not. Their company that is building the system of energy distribution and the car is one of the outlets. So when you think like this it's called natural thinking. The market gets much bigger. So to create those valuable positions you don't think a little bit about it. So that is the ecosystem, right? What you're building here today or starting today or already starting by using artificial intelligence is a new ecosystem a new logic. It's not just a new software. If you're building artificial intelligence for quality interests you change the entire structure of what happens there. Can the software anticipate? Yeah, it's entirely possible. If, for example, I was delayed on my way down here, right? So if that happens, the software can already anticipate the rebuild if you see a person on that plane the very second the pilot says we're not going the software can do that in two seconds. You know how many people take in the call center? How much confusion they cause? But what would be the ideal ecosystem? Look at these numbers. This is a McKinsey study talking about the potential economic impact of disruptive technologies. Each one of those is based on artificial intelligence. Again, down with the slides later but it's automation of knowledge for robotics autonomous vehicles the internet of things mobile internet because it's also something where India can take leadership. This is really, really hard to work here. This is not some of the thermal stuff. There's a lot of infrastructure a lot of engineering, a lot of thinking there and it's a new ecosystem of McKinsey says $30 trillion again. So I'm going to show you this quick video because it explains also that when you're inventing something the timing has to be right and the context has to be right. This is a quick clip about the server-wide process. Introducing the all-new self-driving car. It does the driving for you so you can catch up on the more important things in life. It automatically takes the right turns because we avoid unexpected obstacles. For all, this is red lights far in advance. I get the point here. You can see this on YouTube. The point is the self-driving car may work but we're not prepared for what it does at all. For the consequences of what it does and how it would work the context is caution. When you invent something like this you have to find the context and the context is ready for that. For example, if you're a Tesla and you're rolling out electric cars all over the world there's kind of a problem when there's no place to charge. So you have a great car and you have a lot of your debt that you can finish. There's no context. So the context is crucial and Dave Attenston once a futurist from Intel said that basically as a context we're going to be surrounded and I think the real check of that is really this. Think of a world like this. Think of a world where our brains are being essentially described as a piece of equipment. It's very much a disembodied way of looking at business. And many people that I work with are thinking, okay if we can make the business totally efficient and safe, personnel do away with people and essentially create copies of human processes that are in machines which I would take for shipping business, right? I was in Greece a couple weeks ago and bought the shipping companies and what they want to do is they want to basically have completely unmanned container ships. You know that's not the way 300 people are in a container ship, right? They want to put 5 people on the ship in 10 years. 5 people. It's a robot essentially, right? The machine will be a robot. Now for that I can see the logic, right? Nobody really wants to be attached, right? I can see how that could work. But it takes things by advice especially financial, right? Well, when you have $10,000 to invest and you want to invest and a really safe investment that's pretty simple, right? You can do that. But anything else this is probably not the right approach because you know there's a cartoon that I found the other day that says it pretty well that the cartoon says no, you weren't down though but says the mother to the child you were born. That's a very big difference. It's better to be downloaded from the internet. So for terms of visuality now understanding visuals is a pretty mind-boggling challenge that is just now being solved. The Facebook system can recognize the difference in those two different kinds of jobs would even be hard for you. So here's a question I have for you. Would you trust an assistant or robot as far as it's like this? Would you trust it? If your financial service provider or pretty much anywhere in the world no matter what the advice would be will say well this isn't the smartest machine we have would you do this? Would you have to sleep on an elephant trust in the elephant not to kill you? Would you have this kind of trust assistant like that? Well the answer is what is the answer? Maybe a little bit, yeah maybe something else could give me more trust people don't trust algorithms algorithms can sort of destroy the trust or amplify it or emulate it I think at a certain point that is becoming a major issue in the world that we do how do we actually think do we really want to delete humanity in our world? That is a very bad idea Well it's a bad idea because it will kill the world there's many other problems with it but it will create useless work really what we want with our software when we use artificial intelligence is this right? we want to liberate the planet we want to set people free to do what they want to do in the first place we don't want to take them over like in a way this is what happened with Facebook right? I mean basically Facebook encountered us for a while and now we are empowering Facebook I mean we are the function of Facebook they are using us to create their assets I don't think that's a problem I think we just have to realize that's what it is you know 10 years ago Google used to index the world of information today Google is indexing us they are using us as information so there are things that we have to see there like when we build something we have to think of it like this we have two missions one is technology and the ease of use and products that we create with this and the other one is the human values the ethics, the philosophy, the principles, the values whatever you want to call it a business that does not have ethics is dead I mean we can't just as you know of course technology doesn't have ethics have you met a robot that has value I'm not yet please try I mean we are going to see a private challenge for us in the future so a great ad here from this company called VIF, you can check out they are creating a video engine based on technology and their ad plan is intelligent becomes some of that we can use some of that we can rip off and this guy in 2008 the CEO of RJ said when you create an utility you create something that gives people time and back that's a very important statement it becomes less upon information as pollution and more about information that help people get through life that is a useful statement a utility and this is how Google has become so powerful today anybody from Google knows should be careful I say here is one of my clients but that's how Google has become powerful there has been so much utility every time you turn around it's like oh another cool thing you could do utility, I guarantee you when your software is developed so much utility that becomes like air becomes like oxygen that's the future great article yesterday I had a great technology review with a great picture that I snagged last night in the airplane you don't want this on the last you don't want your technology to trample a client you don't want your client to feel like you're doing this with them like they're becoming a chip but nevertheless your service is a chip it is automation, it is intelligent you want this, you want the client to have this sit on top of the thing that you build and I think that is really what the future is for this it's this combination of intelligence and automation which is inevitable and there's lots of studies of this I think India is really one of the toughest places for this because automation means job loss that's quite clear it will take longer in India because of the size and market and infrastructure maybe 10-15 years but automation kills jobs and everybody keeps saying that there's a new job because of technology but this time it's very radical automation so I think it's actually good news but really what we have to look at is how do people fit into this consciousness I mean there is such a thing as a conscious investor for something and where is sentience that's the connection you need to make when you build he's a figure from IBM that's going to make you think all of the problems in the world can definitely be if men were only willing to think problems of the world all of the inefficiencies complexity bad information bad decisions could be settled easily all of the opportunities could be realized if we were any willing to see patterns in data that we could never see before put analytics in our hands reinvent businesses in the cloud fight cybercrime with math design a machine that thinks like we do if we were only willing to use data and science in curiosity to track epidemics you get the point here right IBM is also one of my clients I have to say this really is kind of strange right because basically it says if we were only willing to use technology which is IBM's technology we could solve everything I mean it says right there if you're willing to think and use data then if they solve it right well the answer is not it is extremely helpful of course right on the left you see a symbol of what's called reductionism and reductionism goes back to a philosopher named Descartes in the 17th century who said that animals could be seen to be fancy machines or he made a picture of a duck that could be copied to the what you call the robot top the machinima and lots of people took this and said well really that's you know animals and humans who are really like fancy machines right and that's called reductionism and I can guarantee you if you're told I'm going to reduce humans to being a duck that's a mechanical duck right it will not be successful reductionism is the opposite of empowerment so it's very important right when if you look at AI based utility right it needs to be holistic this is taken from a wellness brochure it's kind of interesting how they fit together whole creativity open minded, individuals connected you know, chaotic, dynamic you stop what you need to connect with people in a holistic way and that's a pretty hard question that's a very very hard question because it's not the same right it's not like this right I can guarantee you if you invent something based on what's called machine thinking it will be good for a year because it looks interesting right but then what that means so really great software will transcend the machine thinking will make humans more powerful it will do this what I call algorithms and I made up this word and it's called a neologism like algorithms right basically that's the two worlds here's the algorithm the technology and here's the humorism which is essentially what happens between people that is the sweet spot for software that's the sweet spot for artificial intelligence that is to put those things together not to replace the humorism that is the production the very important thing that is mine when you look at these people on the left the so called Millennials Millennials are people that were around teenagers around the turn of the of the of the Millennials of course right so if you were 14 around the year 2000 you're about 30 years old now that makes you the millennial okay and these people are the driving forces of society in the next 5 decades in India actually people who were younger than that also the natives because of the southern population and guess what the you can see a little bit small but guess what the Millennials are saying is the most important thing in their life what they're expecting from service providers the first point the very first point is they're saying they're expecting sites and apps and the second point is they're expecting some form of human interaction which is a very opposite way first point is mobile and website technology second point human interaction and these people will be 75% of business in the next 5 years most companies will be starved by 50% to 70% Millennials just because of how they flow especially in India so they like to continue until the next 70 to 80 years there will be two things I turn this around and I say basically artificial intelligence could also be perceived in the beginning to be intelligent assistants so IA and AI a bit of a role play but basically IA is a mild version of AI so IA is intelligent augmentation intelligent assist all the things that are happening like how you drive a car on the freeway in Germany you can have the car take you on the right lane or keep a distance from the next car that's called assistants and there will be tons of goals made with intelligent assistants so another word that will be a long time before I can go on the German highway with a car that drives itself but assistants is here now so also in your work when you're building software think about intelligent assistants augmentation the stuff that happens here that is kind of the next step and basically definition really is quite simple IA is not replacing workers but augmenting their capabilities wearables, visualization virtual, mental reality intelligent that is the opportunity right now today to have a mild version of artificial intelligence folks are using glasses to visualize information well that's going to become very normal you see here on the left we can hear the right business now this is the growth in digital assistant assistant Cortana Siri Rubenau that's the growth I mean it's huge I mean this is actually way way smaller on the financial time I guarantee you every single transaction in the economic sector with its financial banking insurance will be based on some sort of service with that sort that's already happening but let's skip that part so let's compare this with something to drive here on the left hand side we have the Volvo car trade Volvo has cars that allow you to go on the freeway behind the lead car and just stay in line behind the car and let the car drive like a trade it's called the car trade and works beautifully because it's so simple as a lead car, a stable lead car everybody is in one line it works this here going 200 miles an hour sitting in the back of the car playing a video game or so that is a long time away this is just like a title on the airplane it's possible but would you do that right that's the life or death situation I think the initial part of what we're doing here in terms of the possibilities of technology is the first part is the assistant as an intelligent assistant and let's skip this part because you know there's a great apple of X but AI which is a personal assistant that's scheduled meetings for you I just tried it out there's a software called narrative science that does the same thing for journalists in 1904 10% of what they put out is already written in software like a journalist software so this guy in the company called came show which is a leading technology company he says something very interesting it's really what we're doing here we're alternating human intensive knowledge work look at this curve here I mean the amount of knowledge available to us today is mind-boggling you're going to decide which argument clinic to go to in Kerala 15 years ago you'd call a travel agent or you'd ask a friend today you have all of the information comparison that's mind-boggling you're going to decide where to invest you could study for the next 100 years this is the curve for an exponential of all of knowledge to solve that problem for your customers it's huge automated human intensive knowledge work that's what he described as a goal and as I said before when you have a virtual world you have a world that is based on a screen remotely looking at the speech saying that's 5 to 10% if you're lucky of reality if you look into a screen you're not actually there and even if the screen is virtual now here nice features of Trivandrum I think this is actually this is Goa I think sorry a little bit further up there when you're on the beach isn't sitting right there it's 100% does that mean that the screen with the information is bad it doesn't as long as we know what the difference is there is a difference this is about the left brain and the right brain the right brain is about imagination discussion motion understanding synthesis all the stuff that is essentially more creative the left brain is the engineering brain algorithms logic now here's a good or a bad message for you if you're in the software business you don't have to work with all the right brain because the synthesis of the two is really by the actions that all sits together right Conqueray said philosopher said logic proves intuition discovery if you make the big position about where to buy a house you can use logic but you know that other than you're not deciding entirely based on logic we should not ignore that because that is actually human process that we need to maintain Kevin Kelly said machines are for answers humans are for question do not remove the question with the software amplify the question I think that is really what it comes about also the fantasy that artificial intelligence could have no opinion do you really believe that a system that's based on algorithms and computing and data is unbiased and it's a crazy idea the very fact that you're removing human factors from that whole system is a bias you feed the machine what you get out from the machine the machine can learn something but they have a bias that it's a machine that's the bias of a machine don't think for a minute that this system will be unbiased mutual opinion about investment or whatever it's not but it's still extremely useful so that shouldn't be the goal of artificial intelligence to be unbiased I mean why? our lives are entirely unbiased unqualified, chaotic why would we want a machine that that's like the ultimate truth I'll skip this for a second so the other thing to keep in mind is that ultimately software artificial intelligence makes things a lot more efficient for example I have been proven that if in the energy sector we would use artificial intelligence and sensor networks of either of these to connect energy grids and home heat systems and solar systems we could save up to 40% of energy in Los Angeles now they have just how we should be connect all the traffic lights about 6,000 traffic lights and they could save about 10% of gas in the daily commute 10% of gas so if you want to drive efficiency that is a good business using AI, clearly but it will eventually end when you've made everything efficient what is the next step? when the coal sector has reduced from 100 billion people to 10,000 people what is the next step? maybe lose 3 more people better so you have to also think about beyond efficiency your business plan is to make your clients efficient that's a good plan that's not permanent efficiency is just one step in the value creation that you get for it ultimately what you do in the software now it's following the the path of the experience it's going to the experiment economy so you go from the idea of saying okay the software is a service or a good or not an experiment it is supposed to cost some kind of transformation what you're building when you're building the AI is building transformative tools that will change people's life and that's ultimately what we're looking at but this is the shift of your business plan as good as this is well one more point about the singularity Ray Kurzweil says in 2027 this point here is roughly when one human brain can be matched by one computer in the power of processing 15 years and then he says in 2050 all human brains can be matched that's because of the exponential laws of technology we have to think about what that means from the internet it's sort of a brain match that's kind of the idea that our brains are connected obviously that will be quite interesting I think to me the scenario of what's happening here is that when we're looking at true intelligence we should not look at an individual rural or individual product but at the network the intelligence in the network the intelligence in the connecting fact when you're looking to make decisions it's about being like a rural network like Metcalf's Ball Describe it's about ambient computing we need to get to the end so I'm going to rush a little bit I have a question out on YouTube it's called Screen Screen has a very very interesting algorithm that they have to go up and what they say is they're using a software to draw complex inference to collect data to then say we can infer information from this this is definitely a future pathway to get out on YouTube to manage a bit more time and other companies are building digital nervousness and sensing me basically the idea of saying to collect information to build sensors to deliver kind of the intelligence that you didn't have before for example the environmental systems and so on the cognitive cloud that these guys have done called cognitive scale in the medical field think of the concept of a cognitive cloud the idea of having information that comes along I think we're going to see a lot more from this kind of people social media this is a very rich curve connecting social media sentiment analysis with artificial intelligence I mean Twitter has fallen that path already just acquired a company that does it exactly this two days ago this is a crucial point take a look at social media this company called Cortex artificial intelligence marketing based on social media very powerful too so I'm going to leave you with a couple of thoughts and then I'll get to the questions if we have time first of all Socrates said nothing vast enters the life of mortals as us without a curse that's very good without a curse not with a curse so nothing vast enters the life of mortals without a curse artificial intelligence is something extremely powerful it's more powerful than nuclear power we have to be aware of that nuclear power is bad enough if things go wrong then they go really wrong but India has lots of nuclear power plants and many of them you need allegedly for energy very important to keep that in mind with this kind of idea of artificial intelligence we're unleashing a very powerful thing and we have to be aware that we have to have similar rules and situations we have to think about the black swallow things for example security I mean if you look for a business opportunity security, privacy standards all that stuff is a huge opportunity nobody is going to want to be involved with an intelligent system if it's not secure imagine what that system would do to you I mean it's bad enough as it is with surveillance and as a snowman imagine what would happen if 300 billion devices are connected to the internet you could wash your wallet or your wristwatch, your clothes your helmet, your pilot right? that's a very very big thing I think this is very important for a business social contract and thinking about the unintended consequences when you're inventing a software that changes how people do things you must think about the unintended consequences basically the future will be about both of these things technology and ethics and business values and to me the most important part is that we have to maintain what I call the human imperative the human imperative is what we call gross national happiness rather than gross national product the purpose of our work is to create happy customers that is the human imperative the purpose of our work is not to create software that buys itself I mean we won't have customers, not brokers so human imperative I think is the key to success let me summarize AI utilities are not about replacing humans but about augmenting and empowering them that is the purpose of what we do today 20 years? we'll talk again but it's basically about business super intelligent software I think this is a blessing but generally it is a complex I wish for you to follow that rule in the future thank you very much for your attention I will make this slide available on my website or through the organization probably so your host will send you the PDF link you can download this we also made a video thanks very much for listening for the next session and if you need to ask a question please contact the MS representatives at your desk as you mentioned the harder the question the better we'll take any more questions we'll take some questions we have 5 or 10 questions okay you can tweet also if you want so with the forecast where the singularity will occur is there a dimension of where ethics will start to enter into AI at the same time? yes, well the similarity just to make clear what that is the similarity is the point in time to where computer has the same power that our brain and to where it can amplify itself so that our robot would say okay, I can get more processing power I can self-learn I can pass the human master imagine a computer in 15 years that has so much processing power if you leave the computer alone it will go out on the network and find updates and make itself stronger so it will grow itself very quickly so the point of singularity really is a scary discussion how that will happen we still have time I don't think that is going to happen for 20 years today we are setting the map for this we are deciding what that could be I mean technology is now so powerful that if we don't keep an eye on what technology is allowed to do and who does what imagine if you have researchers researching the fact how to build super intelligent robots that could quite possibly be a lot more serious than nuclear power so we need standards we need I mean we have standards for everything we have standards for bioengineering we have standards for nuclear power we are going to meet that for later I mean and this is why I would advise you take a look at intelligent system IA before you get into the top level of AI there is so much work to be done using intelligence to make augment what we do and it is extremely powerful opportunity and that will eventually become AI but there are very large issues about leaving decisions to machines for example you may have followed the debate about the army saying that they want robots who could decide how to kill I mean imagine if you were an army commander of course you would like a robot that could decide you don't have to call back in the meantime they get destroyed it sounds like a good thing but it is a really good thing how would a robot know this little girl is not a walking landmine another robot for that matter and would she want that robot to decide that and who would be responsible if these are the decisions that we are going to have to face so I think it is a lot easier for us to say let's build the things that will make our lives easier and not that will take over our lives potentially I think that is a much better business plan for time being that is really what we are talking about but we have to be aware that the potential of this technology is extremely powerful and look at your kids there have been many of the kids my kids at least they have replaced some human function with the robot's own when you ask kids between 15 and 25 and most countries around the world who is their best friend you know what they are saying is the robot that is a sad story I think that will change but if the robot is their best friend then you have a real social problem of everything so that is something we need to look at we don't want people to become machines because they use machines that would defeat the robot so I think the best thing you can do is create a software that really does this really well becomes very almost addictive to it but empowers at the same time there are great examples that users already claim don't take it too far but take it far enough to be really exciting for people we as people who develop software typically think about this particular domain so within financial you have financial advice you have investments so we develop software for a particular domain as AI becomes more prevalent do you think the domain knowledge would be the AI algorithm would be domain agnostic as in Watson can do bookery shows, Watson can play quiz shows Watson can do medical diagnosis Watson can do financial advice so when we think of the future in which software and AI become integrated do we want to think of the domain being separate from the algorithm or domain being developed the algorithm developing the knowledge through domain specialist or domain interactions so is the software going to be domain agnostic for a future software application I think the software can be agnostic you can have software they can do all that and IBM is trying to do that for example with Watson he needs the marshall of that software to be extremely deep into the expert check to do the right thing otherwise it could be potentially quite dangerous imagine if Watson would do medical advice for diagnosis without deep domain expertise of all the contextual things that would be because there are so many things that are not fact-based and Isaac Asimov who was a great future president he said one point in his writing that he does not want to be a speed reader machines can do the speed reading he wants to be a speed understander so let the machines be the speed readers right we need to be the speed understanders you know understanding is quite different than reading and as I'm sure you know I mean when sometimes when my clients say they want to know what happens in developing countries and just stay there and then you're inside to understand try to really understand what you're here you look at everything you understand but you can't watch a YouTube video that's not the same thing so I think their answer really is I think software could be domain-positive to a large degree but how we apply it would be largely up to the experts to run it and that will probably be true for at least 20 years we were talking about again the input I get from one person I talked to here they didn't put in terms of data and the benefit doesn't even exist to copy that between you so that's why we should be careful a lot so substituting one for the other we need financial services clearly if we have enough data real-time data social media data object data permission-based data internet-based data and if you have the main expertise you can build something pretty damn quick there's no doubt about that much better than what we have now at this for sure it's like when you look at doctors I do a lot of local doctors some people are saying that 80% of the work for doctors could be replaced by a smart assistant or by a good nurse with a smart system diagnosis I mean simple diagnosis like cold or snagged by it fairly straightforward not about complex things so the doctor in the near future most doctors will have free time so to speak because they're freed up from the manual work like diagnosis medication, prescribing and they can learn new things they can be human again so it becomes a whole different way of looking at what we do a salesman at the bank they're going to say well like the robot sitting on top of this robot it makes me extremely agile and quick but I have to run it so I don't run the robot it won't have value at least that's my thing I don't know I think great courts would have a difficult opinion saying that the computers should take over all of this I don't agree with that anyway, another question we'll have a panel discussion later as well excuse me so like you're seeing this so when do you foresee Google running a proxy government for all the countries in the world what do you do with that what do you mean Google is on its way of becoming the largest artificial intelligence company in the world that is what they do because in five years if Google does not become what's called the global brain the global OS they're dead Google makes 2.8 billion dollars a month with advertising you know how it makes that I go to the surgeon and type in Sushi Trivandrum that is a stupid search to think of that's two keywords of course it's very easy to make a match with keywords that's like saying I have a fun with them give me medication in five years the system will know you you won't type anymore you'll just speak and that is what Google wants to do and that is a very scary thought well I mean I always say basically what's happening here companies like Baidu and Google and others in Facebook they're becoming the new oil companies ExxonMobile and Shell they have to be regulated just like the whole company all they have to self regulate somebody has to take a look at how they do it I mean Google has roughly 2 billion years the internet in five years in 2020 will have 6.1 billion years and almost every one of those new users will be from India and China from the new countries and Google will be essentially running the OS it will be the operating system of information so that's something we have to think also in terms of sovereignty what kind of rights do we have the system that based in Saudi Arabia I'm not sure that's a good idea so I think maybe we'll go back to it but as I was saying I think the solution is not to say that we shouldn't be this because it's going to happen anyway we need to find a way to balance the benefit of the unintended consequences if you have another question or comment I'll get into a deep discussion we could probably talk until the rest of the day but one more question we have one more question cannot build unbiased systems it will be always biased but if we look at some domains we are creating systems for one particular domain and based on a study so how can it be biased I couldn't get to your point by the very fact of what you're putting in I mean TripAdvisor says the same thing go to us to find the oldest opinion of this restaurant but let me take 50 people from Manchester and send them to Malaga in Spain and the people of Manchester have eaten Indian restaurant once before they go to this really bad place in Malaga there's 50 people from one area and they're all right with you that the Indian place in Malaga is the best Indian place in the world so therefore because there's 50 of them the system said well there must be two this place and I have this problem I go to Malaga I go there I have the worst effort and it's because this 50 didn't know that it was the worst it was just lots of numbers so for the system this is the best place in Malaga in general not just for Indian that's not true so when you have you can't say that the data is unbiased just because of your controlling its input but the system is carefully balancing 100 million input feeds and it's valuable and it has so many many more than humans can and the bias is that it's a machine doesn't have any feelings about investing in an arms company I mean the system just says looks for cluster extract so it tells you to invest in market margin like you would never invest in that you wouldn't know that so I think that's something we're very careful about I think it's better to say that it has a bias because it is a machine and that we accept that it's based on the left brain and we don't have to have right brain because that would be really confusing in my opinion I mean I think if I was to advise I would say hey but it's still working that's the interesting part it's still working even though it's not true so that shows you something the system where you say it's very useful because of what we put into it and then you use your own job then I think that's the most powerful system you could have ok thanks very much for listening see you later session and as a token of appreciation I request you to accept this momentary sorry that's a part of kind of some communications ladies and gentlemen guard here now alright so that was our keynote address today we now go more into focus into the financial services domain keep taking investment as an example before I introduce the next speaker I'd like you to see a video over here on investment intelligence