 Wow, we are live. Hey, welcome to this live show here in Corona, still Corona time. Hi, so how are you? I'm great, Nicolas. It's been a while since, you know, we're doing something like this together. But I'm doing good. Thank you so much. Yeah, it's like this is here. I'm just going to open this one. Yeah. This is one of the examples of the videos we were doing 12 years ago. Yeah, I know. I remember this with Simon. Yeah. So this was the VMworld 2008. And it was really cool. We're doing some cool videos. Can you explain a little bit? What are you doing back then? Sure. I think this was back in 2000, late 2007, beginning 2008 when I got in touch with Tom Van Acht. Tom is a Belgian entrepreneur. And you've got to know him as well after that, of course. So Tom had, I think he, Tom met me when we were doing, I was doing a virtualization summit in Belgium in 2007. Virtualization, as you know, back then was a super hot industry. I mean, there was so much happening. Innovation was actually moving towards making, you know, developing big companies. VMware was sort of becoming a bigger company than the others. And so Tom, he had asked, hey, Terry, would it be a great idea if you could go to the VMworld and sort of talk to all these industry leaders and understand from them and then we'll make a video series. And that's how we got started. And I think you knew Tom or Tom reached out to you, somehow we got connected. And then we had some wonderful, I think we had a week there or something. It was great. And what was cool is the video is really long and very technical. And like it was the coolest virtualization videos in the world in 2008. That's pretty sure. Yeah. But we haven't met since then, right? No. Yeah. I mean, sort of, you know, life goes as it goes. And I was back then working for this company called Athos. This is Susanna, Susanna Kirksey, also a wonderful lady. Back then she used to work for ClearCube. Really wonderful memories. Yeah. I think it's, virtualization has been fascinating industry because it was evolving from 2005, 2007 and 2008 it was really getting hot, you know, potentially what AI is getting like, it's kind of becoming red hot. And that was really wonderful to talk out, to talk with all these people and learn from them. Obviously, we were setting up ourselves, me and a few of my colleagues inside the company with Athos. That's a big sort of NVIDIA partner today. We were setting up our own virtualization practice inside the company. It was a big effort. We spent a couple of years doing this. You know, grew from just a handful of people to a few thousand today. I think it's like in many, many thousand people in Athos. So, yes, it's a great way to work. I think that's the thing that I enjoyed the most from this event. You talk to all industry leaders, people developing, doing things. This is Richard Kharsaghan, he's a Dutch guy. I remember he used to have in all these LAN parties and doing VMware user groups and all those things. It really, really made a company like VMware super successful. I think a lot of companies today should learn from VMware strategy where they brought in their services and portfolio into the hearts of engineers. And we were super excited. So, I thought this was a great opportunity to do this, Nicholas. And VMware, since then now it's part of Dell, right? There's a lot of mergers in 12 years. It's a long time. And just because I'm a YouTuber, I'm sure people are looking at the view count on these videos. And actually, that's 2008. That's like the birth of YouTube. And I was putting all the videos on BitTorrent and streaming using Divics on the Internet. So, most of the views were actually on Divics directly from a server. And then also, I was more into Google Video back then. So, there was way more views than what are on YouTube. There was thousands more views on these videos. But the whole video got closed, you know? I even forgot. I mean, now that you mention it, Nicholas, I almost forgot. Google Videos, you're right. Yeah, I was a little bit negative on YouTube back then because it was using Flash and it was like anti-Flash. Yeah, it was crappy, yeah. But since then, I think it's evolved dramatically. YouTube is just, it's insane. YouTube is the biggest website in the world in terms of bandwidth. Amazing. It's huge, right? Amazing. You have close to like 100,000 followers, I see. You've also done well. Wonderful. Oh, yeah. 63 million views on my YouTube channel. Wow, wonderful. I hope it grows. Oh, wow, look at me. This is me with my hair, long hair and tie. So, I was dressed up for the occasion. But so back then, so were you, so what have you done since then? Can we, like for example, when you're linked in, it says something about the WHO and AI. Can you mention some, what is that about? All right. So I think, yes, from 2008 until now, if I start telling you all the things I did in the last 12 years, it might become a very, very long conversation unless you want to do it privately with Bayer, et cetera. So just to, I guess, long story short, after that, I did my first startup. It was in strategy consulting. And then I started gradually moving into this field of analytics in 2011, 2012. So we used to call it analytics, social analytics and all those things. You know, people, you know, NLP and AI and ML was still not so super sexy or hyped up. So in my startup phase, I was trying to create a social analytics sort of a platform, something like what you have Slack today. So it was something which we were trying to build on the back of Microsoft technologies. And there I got excited about sort of, you know, text and trying to understand and putting text and, you know, that content in a vector space. So all my sort of learnings, you know, back in the university, this is like close to three decades ago. I was like, hey, this is interesting. You can apply mathematical principles, simple sort of, you know, trying to get some reasoning out of the information that you have in a vector form and start making some kind of analysis and eventually also some predictions. So optimization was very exciting. So I liked it. And then I think around 2012, 2013, things started getting really serious. I went up and worked for a few years for another strategy consulting form working inside the, well, you know, back then we still called it data analytics. But it was all getting sort of machine learning. So I guess after like 2015, it started kind of going really sort of rapidly. In 2016, I decided, well, I think I need to explore this personally myself a lot. When you work inside a company, there are other constraints that sort of don't allow you all the time that you need to learn this. Let's be very honest. AI is a field that has been evolving for decades since the 40s and 50s. So if you want to think that you can become a master in AI, you really, really have to sort of ask yourself this question very seriously because it doesn't happen automatically. So in any case, I quit my job and I guess, you know, I thought it was the right time to learn. In 2017, I started actually spending a lot of time traveling across the U.S., went into the top machine learning conferences. I had the honor of also sharing and hosting a conference in San Francisco in Silicon Valley. Actually, then it hit me. It really hit me. It was June 2017 and I was in San Francisco. So this is the heart of all those engineers, guys from Lyft, Uber, Google, Facebook, Amazon, all companies, they were their Netflix and all of them, they were sort of deep into their own specialization around machine learning, deep learning, whether image analysis or understanding and making sense out of data, whether text or images. And I said, wow, this is something I need to do. I need to understand this and sort of what is this that these guys are doing that I'm not doing. And I realized that there was a lot. So obviously machine learning, as you know, is a field that's been around for a while, right? But deep learning actually started sort of getting sort of more exciting in 2016-2017 timeframe. So from that perspective, I kind of established my first company in 2017, end of 2017. And then launched a bunch of other initiatives I can go through with you as we speak. Well, you know, fast forward to today. There's been a lot of things. I've spent insane amount of hours. We've trained a little more than 27,000 people around the world in AI, machine learning, deep learning. I'm a visiting faculty at the University in Dallas in Texas. I'm part of the advisory committee at the University of Utrecht in the Netherlands. And then together with a few healthcare companies where I also started learning a lot about medicine. So it's been a wonderful experience for me and Nicholas, I should be very honest with you. I learned so much. And I only realized that actually all the things that we were looking in silos in the past decades is coming together. So virtualization was looking at the infrastructure silo 10 years, 10, 12 years ago. But there was no high performance computing silo, right? The interviews that you've been doing with NVIDIA would have exposed you to all the genomic analysis, you know, trying to understand, you know, even in discovering the, or visualizing the black hole which you saw a few months ago on internet. All this is happening with the help of deep learning in these deep learning technologies. I can go deeper into that if you want. But the thing is it's been for me almost like actually going to a university. I'm back at the university. Probably maybe there is something deep inside me that wanted to go back to school. And probably that is true as well. So in a way 2017 until now, it's been a journey I'm learning a ton of shit. I'm learning on a daily basis. I'm reading so much research. You know, all this, you know, I have my books, all the books that I have here in this room in my study, in my office, home office is full of mathematics, you know, linear algebra, all the medical medicine I'm learning a lot as well because we have many healthcare customers. And World Health Organization was one initiative in which one of the medical institutions in Germany, they reached out and they said, hey listen, I think we think we should try to expand the use of AI and also from a European perspective that AI should be understood. We should be able to interpret it and we should be able to explain it because it seems like a black box to a whole lot of people around the world. And so I thought it was necessary for me to also step in together with my partners in the medical institution in Berlin. It's called Charité Medical College. And together we want to establish and promote AI. So, you know, there's this proper peer review. People understand about the use of AI in healthcare. You know, today the world is reeling with so much pandemic and all these other problems. There's always a discussion around good data, bad data. And even if you have good data, do you do the good analysis or do you do still a biased analysis about the world? So it's very important to put our scientific effort inside institutions like World Health Organization, United Nations, but also advising governments and these services. Because unfortunately, I'll be very honest with you, Nicholas, it's only the tech industry that knows a lot about this. And there's a very limited world. Like I said, in 2017, my eyes opened because what I saw, what people knew inside Facebook and Google, I said, this is wonderful. I mean, they are wonderful researchers. They're lovely friends are having all these companies. But I realized, well, the rest of the world doesn't have that. Will the rest of the world have that ever? And this is where we need to focus on right now, actually in the next couple of years. So there's a lot of talk of the WHO right now with all this corona stuff. But there's also talk about how they're funded. And sometimes it seems that it might be limited in what this organization alone can do. So it's all about what people put into it, right? Absolutely. I think the World Health Organization needs more scientists, specifically more data scientists, but also medical professionals working hand in hand together. So we are not medical experts. I've learned a bit about, you know, especially pathology, when it comes to oncology, some aspects of it. But, you know, it no way is comparable to people who spend 10 years understanding, you know, their own specific domain inside healthcare. These are the doctors and physicians and surgeons and then epidemiologists and this public health professionals are also very different breed of people who learn a lot of different things. Now, you bring this world together with people who look at data and are able to look at data in a very different and an interesting way. Look at factual information, being able to process factual information. It can be a great bone for institutions like World Health Organization, but also other sort of institutions that are definitely underfunded and also the funding, you know, we don't want to get into politics, but it's about using the limited resources coming up with the great solutions. And I think we need to bring in more science inside these institutions. Is it true that the potential of AI in general is so huge, but sometimes it's also a great challenge to connect the doctors and the medical people with the potential of the technology. And for example, scientists also and all kinds of scientists, if you can connect what they want with what the AI can potentially do, you get much more than if just the AI people work alone or if the medical people work alone, right? And how much is that connection happening? How fast is it happening? It depends where you ask where the connection is happening. So I think I would say unfortunately, and I say that with deep pain in my heart, that in Europe, it's not happening at a pace that you would want it to happen like you see in US. You know, if you look at institutions like Stanford Medical College, Harvard Medical College, a bunch of other institutions, they collaborate extremely well with the tech industry. Same thing you will see in China as well. We've worked with a few universities there and a few technology companies in China as well are finding ways to collaborate with each other in a much, much efficient fashion. Does that always work the way either party wants it to work? I don't think so. It will probably not never be the case. But they are definitely at a much, much faster pace in East, in China, specifically in China and in US to develop solutions and products that will help actually these medical professionals do their job better because they are bringing in technology better. So from that perspective, I think we inside Europe will be, I wouldn't say the losers in this bargain, but we become the recipients of these solutions. If we invent them, if we build them based on the collaboration, the relationships that we should be having inside these medical institutions, then I think it would be a great place. So my answer to your question would be, well, it's geographically a different answer. In Europe, the answer would be, I wish it was better. In US, in China, the answer is how efficient and how reliable and how actually trustworthy is that? Because when I was doing all these supercomputer interviews in the past few years and AI kind of interviews, I was always kind of thinking that this was potentially going to really help out a lot when something like Corona shows up. And there's all this talk about the models people have and people using big data to analyze the treatments. For example, there's a big study from Lancet where recently they checked like a million people in all the hospitals who got some hydroxychloric, something like that. I don't want to mention the word because YouTube will delete it. I'm joking. But they're using AI for that. So there's the challenge between what the doctors see when they see a patient and really using AI and then just going through data that, you know, if you get the wrong data, if you put the wrong data and the big data, then maybe the result will be completely different. Then maybe it might not help. Oh, absolutely. So first of all, I think it's important to understand the difference between existing pathologies where we know how a pathology exists, whether it moves into an oncogenesis, meaning becoming cancerous, where we have contained and understood that the pathogen, the damage that it does to our body. In Corona virus's case, we are still trying to understand of its, let's say variability in the mutations that are sort of already there are 26 mutations, I believe, because it's very, very easily transferable from human to human, which is actually very dangerous. You know, today, it's strangely dangerous, actually, if I say it this way, because it's like a flu, but the only the flu that kills you. So if you're going to the hospital, you have an x-ray, which could help you in traditional, let's say, other imaging sort of technologies that have proven to be successful have been reliable and doctors have been able to cure you. Okay, it's just a cough. Or well, that black spot is because you smoke a lot. I had that. I mean, when I was already like in my twenties, they could just say in the lung, you know, the black spot is because you just smoke two months or you smoke, I mean, I guess. But anyways, those are the things that you can diagnose and there is a path to recovery. In the case of COVID, first of all, it's new. That's why it's called a novel. It's SARS-CoV-2. It's a novel virus. It's evolved. And we are still trying to understand the impact it has. So we understand how it actually gets inside a body, how it uses the ACE2 enzyme to actually lock and get inside our cellular system and starts replicating because it's a virus. It's an RNA virus, not a DNA virus. So basically, that allows us to only understand so far as to, well, it has corona or not. But there are the techniques today of diagnosing and then suddenly jumping and making false conclusions and to cure in which you mentioned the, well, I will not use that word. You know, in which are anyways, the president in the US was making some interesting correlations. But I think it's a very dangerous thing to actually right now start saying we already have a cure. That's not the way medicine works, unfortunately. I wish there was a way. To come back to your question on technology, could we have been better prepared? Let's say if NVIDIA had given, I don't know, 200 servers to work with the organization. Let's think hypothetically. Well, it may have helped the process of uniting the scientists together better. So because there's also a lot of division around all of us trying to figure out the way to solve this problem from diagnosis perspective only. Now, prevention and cure, which are two other areas, how to prevent this herd immunity, all these discussions which we're having in Europe, and cure is vaccine, which is going to definitely take a few years. I mean, all these things about, well, end of the year, someone is going to come and get it because that's not the way it works. But again, I'm not a specialist, but this is how I've seen coronavirus where people had discovered coronavirus back in the 50s, so they are pretty old. But this mutation, this strain is creating, well, I can use the English word strain as in a challenge inside our community, scientific communities to quantify, well, is this corona based on X-ray CT scan or even ultrasound imagery? And there are more promising results from ultrasound than any X-ray or CT scan. So slowly we are learning, but unfortunately this is the time we have to unfortunately go through to understand this. So there's no clear answer right now. The only answer is to practice social distancing and use masks, et cetera, because that's the poor man's solution. Unfortunately, very sadly, all this hype about technology and greatest servers who are cracking the code of the universe are not able to crack the code of this damn virus, which frustrates me a lot, but I hope we can solve this going forward. I really hope that the world has learned now if the SARS-CoV-3, which will come, I can guarantee you because this is the way a virus actually is actually maturing and learning about human morphology. But it's understanding about us. It's understanding our biology right now. It's studying our biology right now not only in our lungs, but in our appendicitis and in our internal organs, and there are a lot of ramifications to that. So when it comes next, I really, really hope that the world is united and has brought in technological resources, has brought in intellectual resources and brought in various disciplines together to form a community. I think there should be a task force to actually fight this from a governmental and institutional perspective. I really, really appreciate the work that Bill Clinton and a few other people are doing, but I think it has to be an approach that countries can adopt without arguments and sort of solve this because it'll come back. So to go back a little bit to your point where you talked about being in Silicon Valley, talking about AI at a conference over there, right? And all those big tech companies. And once I spoke with a guy who works at Google who, I thought that they had some kind of, that they would prepare maybe somehow, have something ready for something like the corona, a little bit more like, they're talking about doing this contact tracing through Bluetooth and maybe there's some kind of way where big tech should take more of an initiative in using AI really for good and helping people like tracing where the virus is going, how contagious it is, even helping people getting tests. There's all these kind of ways that maybe they kind of missed the boat on it, but maybe it's hard to blame people because even they're a huge company and they have a trillion dollars, it's maybe not enough to provide a huge solution. Yeah, I think, let's say, if we were talking about contract, I think the technology industry never signed a contract with the governments and with anybody else that we will solve public health issues or something else. I mean, I know I'm saying it very black and white. But I hope you understand. In US, the industry is extremely technology, fast-paced technology, exponential growth curves, technology will solve everything and they're doing a whole lot of work and learning a lot about these new technologies, right? If you look at Facebook research, Google, I'm just talking about research because I really love those people out there and they're amazingly brilliant people. This definitely, so definitely this knowledge, this acquired knowledge of our research and bringing this high-performance computing, understanding how we actually are able to quantify data and actually qualify whether it is going to be potentially sort of coronavirus or not. They do definitely have the intellect, but there is no contract per se, as in there's no unwritten or written contract that sort of gives them the, that creates a bridge between governmental and public health institutions to work with technology companies. I wish there was. I don't think, to be very honest, and it might seem that I'm defending companies like Google or Facebook, but I wish there was some kind of a bridge in which both these things could come together and work towards these problems because inherently I really don't think that people who are working for these companies don't want to help the work. They definitely do. So I wish, we need to figure out a new contract. We need to consider a contract which talks about public health, public well-being, inequalities and economic sort of disparities, creating asymmetrical relationships between us or wealth, between the technologists and non-technologists, asymmetrical sort of relationships economically between people who have accumulated large wealth and they're not sitting, they're not wicked people. They're just normal people doing their jobs and suddenly they got rich, right? So you cannot blame Jeff Bezos and all these other guys of suddenly being bad people. That's not how I think we should see it. But on the positive side, I mean there are some companies like Amazon, they're trying to at least sort of move into the area of climate which is a huge thing, right? Right now, corona is a big problem but climate change is really, really going in at waves where you and I cannot imagine. The bottom of the sea level is warming up, the bottom, I mean deep sea, deep sea, not just regular sea, deep sea. That has huge applications for this planet. Coronavirus is just a warning actually. It's just the beginning. So if I may bring in some comments, Paranjit is saying that the big nine control AI, which big nine, you know, big nine he's talking about, is this situation likely to change? Will it become more democratic? The EU is far behind. We're both in Europe, right? Yeah. What's happening? So I just saw the comment, Amy Webb. So I guess, yes, the big nine I think he meant with a few of the US companies and a few Chinese companies like Alibaba, Baidu, Tencent, and then I guess altogether collectively a nine. That's what the gentleman sort of refers to. I think we, I wish, I really wished that, you know, Europe could learn these wonderful technologies that companies from US and the manufacturing capabilities that China has actually sort of literally manufactured for the rest of the world. I wish Europe really picks up and says, well, where is our European cloud infrastructure? Where is our European AI platform? Where is our European, I don't know, even search engine or social networking? And I think if there is, and I think there is tremendous amount of funding in European Union, but if we keep chasing the wrong goals, you'll keep burning a lot of capital and taxpayers' money and you will never be able to erect, I don't know, a sales force kind of inside Europe or a Google, the next Google or the next Facebook. And I think it's great for competition, believe me. I think as much as people think, oh, it's not going to be good for Google or Facebook, companies will love it because essentially competition will lead to talent growing. So Europe, Asia, Europe and US will have talent all, you know, spread evenly. And this asymmetrical sort of relationship, economic and technological sort of perspectives will hopefully change. So Europe needs to do a lot, I think. And I'm really sort of saddened that we really are not looking at it. And when you show Margaret, she's wonderful actually in doing it, but I really would love to see Margaret sort of be surrounded by people who are sort of looking at this from a technologist's perspective, who are so driven, who are so hungry and who also understand other parts of ecosystems to incubate and ideas and create sort of maybe companies, a bunch of maybe 10, 15, 50 companies of 100 to 200 million Euro run rate. You can do that, but I think there needs to be a very, very significant effort to put this together. So, you know, as being in Europe and the way I look at it, that the web was invented here in Europe, the ARM CPU, I make a lot of videos about ARM is European and it's basically 30 billion devices per year. It dominates everything. It's Linux right here in Finland. And so we invent everything basically. We invent all the tech that everybody uses, but it's also a question of making money, right? And if you make the money, then maybe you can also control hardware. Like hardware is China and iPhone apps is the Silicon Valley or a little bit more than that, but like software over there and what are we going to do here in Europe? We should, the next big thing maybe is doing smart AI. What do you think? I think so, you know, first of all, if we do not educate, there has to be sort of a, you know, in Europe we have 500 million people and there are many, many people coming into Europe, migrating to Europe. I think what we first have to do is before we say, oh, let's just start inventing or building factories to create a hardware or let's invent, somehow let's invent people who can create software. So I think there has to be some perspective. I'll give you a practical example. So I just wanted to share the screen and show you, I think it's already shared. And I hope you can take a look at it. I'll just browse through this. So while I keep talking, so we are participating and this is just a platform we are launching. It's not live yet. So people commented, I know on LinkedIn that hey, the text is still Laura. But I said it's still in development. So what we are trying to do from a part in Europe is this is liveai.eu. So it's a platform in which we are bringing these skills together. So the thing you see in Silicon Valley, people, what I meant is people have mastered these five plateaus. People have been learning so much about software that it becomes super fast. And this is the thing I presented also in Silicon Valley a few years ago and also across the world. I think many people would be familiar with this when they've seen my keynotes across the world in U.S., Europe or in Asia. So what we are trying to do is bring this whole understanding for actually to a few hundred million people in the next few weeks to teach them about the fundamentals because you need to learn this. Universities could have done a better job but we're trying to give them the latest skills on teaching this skill about data visualization which is a second important plateau talking about machine learning which is how the world is actually being transformed. And obviously the most popular baby of this whole revolution is deep learning and this is what people need to learn. And here comes your answer actually, Nikolas. Applying AI is only something today if you look at Daimler and I guess I'll just stop here so you can sort of go back to the to keep this visualization for viewers to watch. So here actually we have to apply AI in manufacturing in oil and gas or in new energy or even if it's manufacturing industry across the world. Today if you see and I will not mention the names I'll just mention the names of a big company, a manufacturing company and a software company in Silicon Valley kind of a sort of equation that we've ended up being in. So there are many many big manufacturing companies in Europe who are tying up with Silicon Valley companies for good reason and they should do that for robotics and AI and bunch of other things. Now because there is not a European answer to talent and because the talent is not able to convert their talent into ideas and solutions and then those ideas and solutions are not able to sell to industries Europe starts looking at okay so how do I optimize or speed up my operations and the best way to do that is to very quickly go on internet and you'll find that most of the companies are funded in US and manufacturing capabilities in China sometimes things are together so it's very easy to buy that I think Europe's answer should be to focus first on the learning it has to be super important that the people around the world are in Europe we have all these hundreds of millions of people here tens of millions of students in Portugal and Spain etc. we should provide and raise the bar so that education and AI becomes a European let's say prerogative something which each and every European student and even around the world the whole world is welcome and to do that we are doing it we are participating in a project I will not go because it's confidential right now we are participating in a project from European Union to set up AI across these tens of millions of people and that is a project in which we are working with the universities in the Netherlands in Italy and in Poland and I think also in Finland and or in Denmark which we haven't sort of figured out but this is some information I shouldn't be revealing more but again this is an initiative which we are setting up to set the education first so people can learn and don't have to say oh god I don't really understand this well I think I need to do a PhD at Stanford or in University of Toronto which is you know the smartest guys are sitting there and the ecosystem has already started evolving up there so that's the first step and the second step is definitely I think providing the ecosystem you shouldn't just say blindly oh you know it's market dynamics and companies should figure out themselves no it's not going to happen talented people will keep leaving Europe because you know if there's one smart guy or two smart people sitting and thinking of an idea but are not able to realize it and someone makes them a phone call and says hey listen there's a great job at Google or Facebook that guy is going to not think twice because you know all the beautiful Europe no offense I mean I love this is this is our home ground and home turf but it's not going to help the people so they believe this is what is leading to talent drain because our bureaucracy is pulling us down so if you were able to create this ecosystem and carve out from that 100 billion euro which Europe has for AI and just give maybe five billion to an initiative I think it's very easy to set something up and create an ecosystem it's it's a matter of choice I think European Commission should really be thinking about this because eventually if talent leaves your country what are you left with you're just left with monuments you're left with Europeans Americans and Chinese coming and visiting the flower garden in Amsterdam and and all you're doing is just I don't know servicing some other industry this is how I think Europe should really focus on it to not to lose edge and at the same time becoming sort of you know equally competitive against these beautiful sort of innovations that we see in China and America so what I'm saying is these are wonderful lessons for Europe Europe should learn from China because it's done all the awful awesome things somewhere my apologies awesome things in manufacturing there's so much wonderful things China has done in terms of learning how to manufacture this is focusing on manufacturing please so this is a technology discussion and wonderful and awesome things and awful awesome in US I'm just trying to make the distribution in terms of statistical statistically making distribution so you can then correlate me and then you know you won't see me calling one country something a lot of software development all these platforms databases AI technologies and tools like TensorFlow fly towards and if you can learn from both these you know beautiful sort of Chinese and American inventions because that is what it's all about Nicholas it's not about it the politicizing is a very different game but the real thing is we really know manufacturing is in a wonderful capability China has US has a tremendous capability software development and innovation and Europe needs to recalibrate and reconfigure on these two aspects maybe Europe can also do something in terms of applying this AI and real benefit to the society right if we can there's so many problems in society so many people have problems with for example work a lot of people losing their jobs right now there's so many challenges in eating healthy or managing resources being happy making sure people can be happy and not like sad or something that there's so much potential to apply technology to make Europe better and actually Europe is the best place to live right China is awesome but I'm joking China is great and America is great but Europe is the best I'm joking but a little bit maybe there's something that needs to be done and sometimes the money people even the politicians want to see results if it's possible to get to those results fast that would be great absolutely I think as much as I mentioned that what Europe can learn from our eastern and western friends both these neighbors can also learn a lot from Europe in terms of the social system the healthcare system that is under tremendous stress and Europe also is going to go through some really really painful things even starting from ourselves my company we are going through some very very difficult discussions but in the face of adversity I really hope that we are also we are working with our customers and hoping that we can pay the bills we can make sure that things can happen but it's a very hard time there has to be ways where Europe can create an ecosystem and we are not going to go print free money I think an average European is a hard working individual who wants to really work hard to make things happen for his family and ecosystem but I think we definitely need to bring in more elements of that innovation hardware manufacturing capabilities doing things on our own and that needs to come back to Europe so it shouldn't be like Europe is a big bus but it doesn't have the tires and the engines and the air conditioning because someone else is manufacturing it because those are the experts so essentially if Europe is that let's say the exoskeleton of the beautiful bus without the essential elements which I am unfortunately sort of think that is the case then it will get hot inside that bus a lot of people will start getting angry at each other and that's what's happening right now within countries and that creates sort of fractures I think the best way is to start providing each other the means so one neighbor can start manufacturing tires the other neighbor can start figuring out on electrical parts of this bus the other one is focusing on some wonderful environmentally friendly air conditioning and then you have the bus running as well I think that is what Europe really needs so as much as you say Nicolas that yes it's a great place but I really worry on the future of Europe in the next 20-30 years because eventually when money dries out you fall prey to a very dangerous situation where economically powerful countries and regions can start influencing your policies which is already happening but again I don't want to go into those discussions hopefully the post-corona economy will be boosted by some great ideas so what you were showing here live AI can you expand a little bit so is this a teaser on what's next what you're doing so live AI is a platform which we have built so we are going to bring actually live meaning this is going to be me like this we are doing live so we're going to have full program of these training programs coming and we'll be making it available for people worldwide the other initiative with the European Union and we hope to find a way to also connect this with that ecosystem will enable that this will be possible for everybody and should be you know essentially any student should be able to follow this free of cost free of cost so that's what we are working on so you know we are able to also convince the necessary stakeholders to do that so this is one I wanted to show you the other which I think the healthcare is also a great initiative maybe let me just try to share that screen as well because live AI is a wonderful initiative and I think people should definitely keep an eye on that and we'll release this platform pretty soon and let me just see Chrome tab and I need to show you an example as an inspiration so for me I was pretty bad in biology back in high school and I studied physics and then eventually astronomy or you know critical aspects critical subject within nautical sciences and biology was a super bad topic so this is for me actually trying to inspire anybody and everybody that you can start you start learning more about healthcare and start finding you need to be a so you can apply AI this is an example of applied AI that's why I focused on that applied AI part the final pinnacle of our platform which we are going to launch soon but this I learned myself I worked together we have a few medical professionals inside our company and we made sure that we take all the sort of expertise from medical professionals learn from our customers but also keep developing and this is also a labor of love I learned a lot myself you know while learning these technologies you know we are NVIDIA partners inception program members and you know working with leading German and also pretty soon we'll announce a few other medical institutions which we are doing this so it's an example and I think you know flipping back to the applied AI this is an example what you can build with when you have applied AI skills so that's why I feel education is probably the most important thing we need to focus on and now as you can see COVID-19 I can kind of quickly kind of show you it says COVID-19 task force yeah so this is actually also awesome we are right now in the process of writing a research paper and also we have a tool ready I do have a tool actually but I cannot show you now or maybe can I if this is something there are a lot of previews I think it's work in progress I need to I'll kind of show you what so we are building tools and there is this tool was built by a developer actually in Daimler can you imagine a robotics I'm just looking at URLs so I'll provide that the car company Daimler yes so a robotics expert he went ahead and he sort of he figured out a model in which the herd immunity how can we kind of make this more possible for people so how to reunite the populations around the world so let me just actually give me a second if you allow me I want to give you the right URL because I think it's worth to show it's important for people to show that within these few years I kept you know persisting and also working I mean not everything is coming to myself automatically there are people who are neuroscientists we work with all kinds of people with wonderful skills and they're the ones who are actually sort of learning a lot from each other and that is fascinating I can tell you Nicholas it's probably the most exhilarating experience that I have today about AI because it's coming through hard work it's coming because we are spending a lot of effort let me see if this is the one I think so it's so it's important to understand the why let me just go back yeah if you load the the URL in the same browser window maybe you can just load it there like that so yeah so you've been busy with coronavirus and AI also oh absolutely so this is what we are working on right now and I just need to find the right URL and this has been developed by this wonderful wonderful professional here you go so I'll obviously have to hide this link later because otherwise people will get too so essentially you know this is a tool developed by this wonderful professional totally from a totally different field but the thing is they have scientific backgrounds people have learned robotics planning algorithms and a bunch of other things people understand what epidemiological models look like you and I can also pick this up so right now this is looking at this is just sermon data and here you have for example if there are no restrictions in this case there's optimization so the daily adjusted restriction of contact without overloading the healthcare system focus Nicholas around understanding how do we reunite the populations you know today everybody sitting and trapped inside the poem don't know what to do the only poor man solution is to wear mask and be social distancing unfortunately and this is where you start quantifying all these things with data so for example if there was no limitation you can see this red line is the places in the ICU and here you will end up getting into serious serious trouble and there's a serious economic damage you see the amount of people who will die I mean that's a huge number this is just Germany so you know close to 70 75,000 people that would be have been dramatic and containment measures I will not go into much detail so this is a time span you can see flattening of the curve was another sort of discussion which you were having but you can take the simulation parameters and actually start looking at how do we sort of you know do that if the period of is simulated to you know if you just push it a little bit further optimization and modeling parameters is also something you need to look at so these are the kind of things that we have built and the most important thing is through these tools and I will not go into much detail right now because this we you know you're getting a preview all these people they're seeing getting the preview today so you can see the economic damage as well so a lot of information today is possible if you quantify data properly Nicholas so this is also what we are working on right now we are hoping to release it in the next few weeks and I'll just make sure that I sort of change the URL later after this chat so but in any case what I'm trying to actually say is that there's a lot possible if we unite scientists bring people who are who know and who want to do use AI and all these modeling for social good imagine why would any Daimler or you know there were people also from Lothanza who are also participating in this project why are they interested in because they inherently believe that together with a safer climate understanding and bringing science and all these other various disciplines who we do not master together we can provide quantifiable reliable and factual information and based on those factual information there could be an interesting model for Germany to reunite the population based on parameters meaning small village you know there could be some other principles of social distancing applied a densely populated area in Berlin or in in Amsterdam might have very different principles that may surprise a lot of people why are you doing this but it's all based on data so then people will see well the infection rate has actually constantly dropped down so those measures are good I know a lot of people are doing that already so RKI which is the German sort of Roboth Koch Institute is doing that in the Netherlands we are doing it in in Denmark also this started much much earlier which had very positive effects for Denmark, Sweden unfortunately you know they made some really sort of serious errors in their judgment if you ask me but I think those are the kind of things we definitely have to focus on and make sure that we are you know making this a reality and this can only happen if we are you know solving the problem together because eventually no matter what you say or you know what I believe if the earth is not sustainable if diseases are going to sort of you know hold and screw the economies this way no amount of wealth and technological prowess locked in in pockets in Silicon Valley or in Toronto or in Montreal or some other wonderful place on this earth is going to be of any use what you cannot take those dollar bills or euro bills and eat because if the whole industry fall apart then and unfortunately I hate to use this but this is where we ended up being who would have thought that this world would end up sort of being totally upside down and but you know ending in a positive note I think the spread and the use of AI in the most responsible manner Nicholas is the only way to solve global problem we will not be able to solve the problems around the world if we are not going to sort of focus on these distributing our knowledge and our learnings and our products and our services in a way that people do not become recipients as I mentioned if Europe is to become a recipient then Europe will end up in trouble because it will only start regulating because there's no other way you need to understand what you're buying if I don't understand what I'm buying or if I don't understand what my citizens are using then I have to find a way to punish you on things that I think not right but the best solution actually that's a negative way of looking at things the positive way is to why don't we develop our own economies why don't we start building more technology why don't we use AI within Europe in a more democratic way and I think all these technology companies are more than willing to participate in that so what we're showing is basically a way for people to adjust the modeling a little bit in terms of this whole corona and does that is there a lot of AI in the way that's built that solution so no this model right now is so we are now right now training this model so the model is based on what the epidemiologists use there are two different models or two specific models people use they are SIR and S-E-I-R so susceptible infected and recovered and susceptible exposed infected infective and recovered infective is the one that is infecting others so to make the distinguish the distinction so these two models so this model is based on S-I-R which has some extra sort of parameters for the control elements such as social distancing et cetera but this model once you know we have to now we are collecting data across a few other countries because you can not just use German East data but we want to bring in Netherlands and other few other countries we are hoping to to release the tool in I guess I hope in a few weeks with some optimization which government or the whole world can use basically that's the whole idea I mean I'm not going to be doing anything big by just sitting on top of this tool doing nothing our research people will also look at the data quality and how people are using data in the right way and while you know let's be very honest will a prediction will there be a prediction tool that can predict some sort of an event I don't know but it can make some probability around well based on activity inside that region with 500,000 people that live there and migration patterns and metapopulations for how people sort of mix around because the world is extremely mobile as you know today you can make reasonable probabilities and they are saying well it's good to not keep let's say shops open till Sunday's 10pm for you know it's good to be it will be a tool that will help governments but also municipalities make some reasonably reliable decisions on being able to apply control in the most democratic way and not just lockdown, lockdown is one way lockdown means everything is shut down everybody gets screwed and that's what is happening so many businesses are going to die so so bad hurts has filed for bankruptcy all these you know Lothanza is getting an injection of 20% from the government what 10 billion euros this was not the way it should have been I think if we don't use science and technology we'll continue to keep shooting in the dark do we need to do that I don't think so lockdowns are basically quarantine we don't have the method we used in the middle ages right and this is pretty much what we have to get back to because maybe we didn't have all the AI ready to help us better in this situation do you think that's correct partly so from the novelty perspective of the virus it's probably a bit too harsh towards our healthcare systems but the governments to say that's so stupid and medieval that this is not like the black plague or something else but they did practice the similar principles back then as well the social unrest that was happening was also similar if you read history it's very interesting but yes I do when I say partly the positive part is if there was a unified dashboard where they were like 500 scientists who had dedicated a few hours of their week just a few hours of their week maybe two hours per week together collectively providing thousands of hours in a week from all over the world fighting this together we wouldn't have problems of blaming labs like Wuhan which is all unnecessary drama because people have been trying their best to understand the virus pathology there we wouldn't have problems we would have had reasonably advisory towards whether countries follow or not is their decision but at least there should have been some global maybe world health organization could have been the primary distributor of this beautiful dashboard explaining things in the most wonderful way today information is coming in thousands and thousands of images and videos and going through all that information trying to figure out which is the right path is very detrimental I mean all these other diseases all these other medications which some presidents of countries are promoting and those discussions will go away because then the world is looking at this dashboard which is created by every scientist from every nation I wish there was something like that because if you don't solve that Nicholas believe me four years from now you and I probably will be sitting again talking about some corona from some other sort of mammal which has jumped on to humans or otherwise or maybe children today we are sort of today you somehow we are looking at the society where you say oh he is 55 plus he has some underlying condition those people are too bad it's believe me that's the kind of mindset too bad but it's not affecting anybody else but it will come in different the strain as it mutates further will start showing itself in the biology of little children in different ways it's already showing by the way in some if you read the news it's going to affect very different populations as well today that seem to be protected because mutation that's what mutation does it keeps learning when every time it touches new tissue hopefully let's really hope that there isn't a second wave that somehow it doesn't come back but it's very important that we need to be ready whenever there is a next corona right because the bats are still in the caves and the pigs are still in the farms and stuff and so there's probably going to be other viruses and hopefully we can figure out a way to not have this happen next time this huge gigantic society turning on its head but on the other hand it might be a positive thing to rethink what society should be and that's what some politicians are talking about a little bit like right like let's not do the same mistakes as before and let's try to be let's say people are a little bit afraid sometimes that their privacy and all these cameras everywhere and China is putting a lot of cameras everywhere but this they also have 1.3 billion people and you need to manage pollution and resources and health of all these people is not an easy challenge so I'm not trying to say that it's a great thing to spy on everybody but if there is transparency and if people can see that there is a point in having your Bluetooth always on and letting Google or Apple help the big AI system you know like use it in a positive way that people probably will be fine with it right and it has to be a positive way yeah I think definitely the most important thing is to understand especially when you look at in this specific case of corona I think we have to be careful in distinguishing between when you need to use let's say positive surveillance means right to bring down the dramatic and pandemic effects of a natural this is the natural disaster in a way right on a global scale so I think you need to make a distinction between situation per situation basis and it's up to countries how they want to use these measures when pandemics come and go in how much will the data be used in the most wise and the most rightful way but it's important to make the distinction that in emergencies for example in emergencies you have lockdowns you can have cops and police and military on the streets making sure that there is no disruption like a curfew but then to move into freeing society and opening up society means you have to allow mobility in a fashion where people don't feel that they are controlled but again I mean from a European perspective it's a big discussion but that will lead to a European based solution I'm hopeful of that in US their laws are different and how we look at public space how we look at large groups and controlling and managing or facilitating large groups in countries like China and India potentially it's going to be a lot different and probably will remain the same as we see today so it's not about pointing a finger at a certain nation and saying well they are doing bad and we are doing good it's about what applies best to us so for Europe I think it makes a lot more sense to take some measures that may be unconventional and then withdraw those methods once you have reunited the populations reunited and recreated and let's say started the economies so whether small companies like ourselves or big companies don't end up failing and dying and in US the move is totally different where they are a little too eager to push and pull the economy so the resurgence will come it will happen I mean if you look back right now you can kind of imagine it's almost like Independence Day if you remember that movie with Will Smith so coronavirus is more like those the mothership is somewhere and let's call the mothership the big mobilization deforestation globalization and sort of how the world has become today concentrated in big cities and the world is supplying foods and supplies and water utility into those big cities and then you know the whole system starts actually becoming very super asymmetrical and then you have those so the first explosion happened in Wuhan and then you had happened in Lombardi and then you had New York and now you have in Sao Paolo in Brazil the next was going to be you can start already predicting it's going to be Africa or other regions there's going to be large quantities a large number of lives lost unfortunately and then the resurgence will also find it's almost like waves it's going to come because right now we don't have any physical or technological or unfortunately even scientific means to to block it physically technologically to invent something, create a vaccine and scientifically to predict it because there's no one out there right now and we're all trying to do what we're doing and this tool I just showed you but there is no conjoined effort in which we are trying to let's say predict it so my prediction is that it will definitely come in resurgence it will last for the next two to three years just like mayors and SARS the first virus in 2003 was here and it was in different parts of the world so the first SARS was in Guangdong in China the MERS was in Middle East from the camel so it's definitely going to come as a resurgence the pathogen will is now amongst us the dangerous part is unfortunately pathogen is now amongst us unlike the previous cases in which the pathogen disappeared because you either killed this large population of chicken or you isolated the camels it's much, much bigger now this one is way bigger than those other two so even though maybe the other two didn't really come back as much maybe this one might have a bigger risk of coming back because so many millions of people actually got it there's too many people with asymptomatic people walking around on the planet the virus needs to continue to replicate and continues in its path of destruction destroying cell by cell how it damages the cell and then dies with it it's going to continue so we hope from physics perspective you actually expect that there's the first patient discussion we've been having so you hope and this is where I mean the prediction I wish if we had a model so we could find and be able to predict the last patient meaning that's where we know that we have well if the vaccine hasn't then the nature has finally killed that last pathogen inside that I don't know 892nd millionth earth citizen so it should technically it should die because it just cannot survive it has to infect your host and just sort of has to die with it it's unfortunate but in a positive way it doesn't live inside the body but I don't think we will be able to find the last patient and until we don't find the last patient we'll be in these waves so it's like a mysterious way it acts it's even the top scientists in the world have a hard time explaining why sometimes these viruses just disappear and sometimes they appear and they disappear they don't get transmissible as much as before right so there's some kind of people thought it might be the weather but as far as I understand it doesn't really have to be the weather it's some really strange way and let's hope it doesn't go to Africa because that would be horrible but let's hope would you mind if I put in a question from Shravani she says what might be the future developments in deep learning a lot of stuff happening with natural language processing image analysis what might come into play what do you think are the future developments interesting and a very good question these are also from I saw an interesting article which Jan Lukun has been talking about called I think I have it here as well I can see live comments I see these comments here myself you see it in the thing you can pull in the ones you want to answer how do you pull that in maybe you can't I'm not sure because I started the broadcast you can tell me which one coming back to this question first I think self supervised learning interestingly becoming a good field of discussion within within AI experts and Jan Lukun you should look up self supervised learning that seems to be has some potential I think the next wave will also be and this is just me based on the discussions that I see on Twitter and Reddit and hear from people the next wave will also be in lifelong learning and one of our researchers is going to be doing his PhD at the Mila University Institute in Montreal lifelong learning will be also very good it's moving more towards the general AI I don't think it will be the AI which people think AI is it's far from that so I think self supervised an evolution of reinforcement learning it will be called something else maybe it will be called self supervised learning or something else I don't know is also something which will play a bigger role I think the application of from the application perspective I think I see a guy saying fake, fake, fake Adrian so I think the application of AR and VR specifically with 5G technologies where you will see that many of these applications today that you're building with computer vision or you're building with NLP will be interestingly in a very beautiful way you will be able to create products and services with that so I think from that perspective the edge computing principles and applying AR, VR and basically applying your machine sort of technique in computer vision whether it's image classification, image filtering and all these other things will be very interesting so I think in the past 5 to 6 years you've seen a lot of interesting theoretical concepts being laid out and you know there have been so many platforms our platform will also go into that but in the applied AI this is where I will bring these 3 bridges together so the research is one field. Then you have engineering meaning creating and machine learning pipelines which Uber and Netflix and all these companies are doing in a wonderful way and learning and creating new sort of ways of data engineering is the second and the third is web development and application in AR, VR so from scientifically self-supervised learning evolution of an evolved concept of deep learning will be something interesting, lifelong learning as well and application wise I think AR, VR will really start seeing its it will start growing you know all these years people have been yelling AR, IoT, VR but now with 5G I think you will start seeing applications and even startups, new startups coming in this ecosystem so it's a very wonderful area in very very exciting times to be in an AI right now so the last 16 years I've been to so many trade shows and in 2012 actually I was CES and I was showing off my Copin Goldeneye smart glass Sergey Brin was just walking around and he was trying my prototype on that Copin was making and I was such a huge and I still am huge enthusiast in the potential of smart glasses and augmented reality but maybe the issue well the number one issue I think was basically this money problem or this decision making problem that big companies have like Google where you know there was a few journalists who are like I would say Apple fanboys and they were trashing it a lot and saying oh you look stupid and stuff like that but maybe the real main issue was the killer apps were not there yet so maybe for this to really work we need some killer apps and those are usable AR applications so something really useful like if you can just wear these go in the supermarket and it can recommend what you need to pick from the shelves to have a delicious and tasty and healthy lifestyle and maybe also help you navigate you in the world or interact with people at the conference in all kinds of ways so it would be nice if those killer apps are coming are you working on something like that so I think I mentioned briefly we are setting up our AI design lab in Paris so we will be focusing on AR, VR there is one of our partners from Russia is a wonderful entrepreneur with really experts in that area collaborating together with them and also a few other companies around the world so in the design lab actually what we intend to do is bring and create this ecosystem where creative people technologists and researchers can come together in one area so to make it easy and simpler to understand so yes one is yes there are no killer apps second of course Google Glass was there long ago but it never found its real application because people were scared there were no regulatory ways to control it but I think the best way is to show that you should be able to contain the data transfer data transformation so transfer and transformation meaning creating a model and algorithm within that limited pipeline to be able to show the productivity gains Google Glass became very successful especially when you're walking around in manufacturing plants so and the thing is taking these apps into public spaces depending on which region which part of the world you live in can be problematic because it's basically this example so as I get older I need to wear this it really is true I have to wear this because I need my glasses to be able to look into the letters that I cannot see but the minute you change this discussion especially when people are not fully informed about the benefits of AI then they will say they are just like me so we like to use glasses because it helps enhance my vision but if you tell them whether in any other profession that a computer like the example that you post can enhance not only your vision but it can optimize when you're walking in a store be able to do grocery within a matter of seconds so basically you spend 5 minutes instead of 2 hours it becomes a very sort of sensitive discussion because we are willing to believe that we control our flaws because lack of sort of gets you get older you cannot see well but we are not willing to accept the fact that AI is doing the same and more so I think this discussion basically needs to be taken into context where people are sufficiently advised around the use of data in a secure manner creating a decentralized AI kind of a model in which you can preserve the privacy so you can still maintain the benefits of this person who is using this glass or this young girl who is on a bicycle going after party at 2 am in the morning because of the glass there is so much safety that people are aware of it and there is no some weird guy jumping out of the nights especially which happens a lot to young women sort of you know and that way you can protect so there are a lot of benefits to technology but it's very different we are living in a very polarized world let's be very aware of it and in a polarized world the divide between technology and the ones who don't understand is becoming bigger and I hope we solve that first at least in our own way and if I may bring in this question as a way to ask also what you are doing in Paris so it's Darshan says the Netherlands is not investing enough in AI and how about Germany and France within Europe is that true? I think so yes we are unfortunately spending a fraction of what Germany and France have announced it's even in the last year's conference I remember there are a lot of AI alliances but they are so extremely cash poor in the Netherlands and you cannot hide from the fact that you have to fund our company for example we are a really small company and maybe I don't know if anybody is listening but if I had some institutional but also some funding of 10 million or 15 million euros I think we would have been able to do 10 times more work we would have been able to create and retain people and be able to develop a lot of things and this is just a fraction of what the rest of the world Germany and France are creating but having said that I'm not sure if Germany and France are doing the right things with the investments we don't know it's sort of a macro figure when Germany says we will invest 3 billion I think that's what they announced in Berlin 2 years ago in AI and no one knows where that money goes so yes it's true that the Netherlands explicitly has never made it explicit how much they invest but at the same time we should think about when countries say that they will invest 10 million what does it really mean is it a political statement is it an institutional statement or is it really trickling down into startups like if someone tells me the Netherlands is not doing as well I think I know because you're trying to survive on our own here and if the Netherlands doesn't do that the big problem is companies will start making different decisions that's the big risk so you talk about this in Paris and there's this talk about 100 billion for AI in EU, there's a talk of 500 billion post corona investments there is potential that the right thing is going to happen next couple years that maybe the EU is the place where solutions happen in AR that actually become really relevant to millions of people and solutions happen in terms of healthcare AI that solves the next corona and that also solves a lot of other healthcare problems in society so this is happening this can happen I hope, I'm hopeful Nicolas, I really hope that this happens because so yes I think in a way you said a very interesting thing because yes, in manufacturing maybe I don't know do you want to build everything grounds up maybe yes, maybe you know I mean look at Tesla if Tesla can just come into your just in Braunenburg close to Berlin and start building factories Europe should be able to also come up with that and start sort of reinventing the way they want to manufacture I think they do that as well but not enough software development is definitely needs to happen more there's not much happening I mean there are definitely some companies who are doing some good job but is it at that similar scale with China right now and US is doing no way yeah, but having said that AR, VR could be the moment where Europe can seize its it could still be that when everything circles around and comes to after 360 degrees Europe can still strike but again like I said Margarita and all these other people they definitely need to be surrounded by people who are builders who are creators who are spending insane amount of hours I mean if you know Margarita you can send her a note from me that I'm happy to help her out she's definitely watching this video I hope so but I mean this is very important and if it doesn't happen companies will perish I will be looking for a job because I don't have anything else to do I have to feed my family so which world do we want we want a world in which Europe thrives invents, creates and sustains or you want to be a world in which we complain the different kind of aims instead of sustain you complain and then everything becomes a side discussion so I want to end positively if you allow me to focus on spending a right energies and maybe creating a European model and I think a lot of people have other views on why Germany and France is coming up with 500 million there's a big, big reason behind it and I think the best way I can sort of summarize this is what my wife said she said this when I was complaining all these countries and some countries are more some countries want less she says the reason and just look at it from this hypothetical perspective the reason why we are friends together is not because we like us each other the reason why we are friends together is because we don't like each other in Europe we have to keep the peace there's no choice there's no alternative than peace so that's why the EU has to be just stronger and stronger we don't want to have conflicts with the UK and we don't want to buy your jelly and you have to buy our BMWs exactly the only way is to get out and come out of it authoritatively as a technologically progressing union as a technologically as a socially we are our value systems are our background and this is the fundament of this Europe the foundation is already here we have to start supplanting those beautiful boxes like the heat sinks which you saw when you were seeing the the GPU DGX1 server each heat sink is like those are the foundations of European society technology infrastructure, cloud and all these other wonderful things and once we have that then you will realize that we are walking in a beautiful garden and we are doing wonderful things and not interviewing other companies and then this new EU commission is recent and the next step they talk about 7 years right is it called the horizon or the new kind of like research investments and everything but needs to happen faster than 7 years we need to have solution so quick the EU is going to have trouble if they can't show results quickly people are going to start losing their patience absolutely you want to impress those complaining Europeans and make them happy be happier it's the most happy population but still there is so much that could be solved absolutely this is again there are two more new questions maybe you want to I can also address these but to this point to your point first very very important that we shorten these cycles of incubation to productization of every other subsidy or grants that every institution and other companies we have one of your grants we are extremely proud we are seen as an innovative company within our region here in the north of Holland but my goal is not to wait 3 years to build a solution which is typically the grand cycle as you know but try to shorten it to less than a year because this is what will make the European Union competitive European Union should have windows of 3 years not 7 years of these 100 billion they should say 3 years 100 billion the next 3 years because the return on investment will be so huge the next 3 years will be probably 150-200 billion and the next 3 years will be maybe 300-500 billion because the return is huge companies became so fast they grew and they are establishing big sort of platforms so I think it has to change dramatically if it doesn't then we are stuck in the same problem of 7 years and some horizon 2020 projects and I don't want to get into those details can be long laborious don't do anything, eat money everybody is just eating project management money and then the money is gone that's not the way to innovate that's writing a certificate maybe this should be the code of this conversation do you want to write code to make innovation or do you is your code an obituary maybe that's nice, ok this question there he talks about grassroots around climate micro generation by the jisters so I don't know Rory if this is a question you are asking when does maybe do you mean when does technology disappear into the background we should reinforce grassroots solutions or you are saying that yes I think it's a statement you are making absolutely makes sense I think first of all technology that magic of technology disappearing in the background needs to be managed and regulated very well because there is different ways how you mean and what one means that technology disappears is very different regulator will look at it very differently than you and I but I mean I am a technologist yes it has to disappear if there was a better way to code and that's what we will show in the design labs copy code from one online and then create an app just by pasting it on a laptop so we are coming up with some interesting ideas to do that that's just AR VR shit so yes I think the grassroots solutions we shouldn't look at it as sequentially Rory we should look at it that we have to actually build grassroots solutions so I think it makes sense what you are saying and Johnny is asking regarding your statements on the software development if you were at the bottom of the mountain again where would you focus your learning towards engineering interesting question so I think first of all you are not ever at the bottom of the mountain you are essentially on top of a certain hill and there are many hills and that's how I would like to look at software development so essentially you can see almost like a convex surface a manifold in which you have several hills so software development I think you have to focus so I would focus on trying to iterate and keeping the focus on AI or machine learning in a way that you can actually do some visualization which may not be a solution eventually but you have actually you are learning the engineering aspects of your iterations so you do first iteration and then think of it as my first iteration is to be able to identify information that I have in this data which obviously is not a solution right it's not a software but iteration and eventually engineer it combining some kind of you know other skills in software development they could mean anything I mean presenting information, visualizing it creating interfaces that are you can use mouse over you can just move your mouse over and then it already is doing interesting things that's a great way to start understanding how to start thinking of engineering when you are actually engineering it into a hardware solution or maybe engineering from a software solution perspective so I would just look at each iteration at least that's what we try to do this is not just me sort of making it up here that's what I try to push my teams internally that iterate and I want to see something I just don't want to see you know hundreds of lines of codes and ten you know fifty python or r or some other files show me if you can kind of you know stitch this together and that's what essentially engineering is right you're stitching components and you're sort of bringing these together that would be my advice but I hope I could go deeper to that and when I look at your LinkedIn which is very active you're posting a lot of stuff there there is this right here this is what is this website yeah so obviously deep kappa is I can also open it up here so this is the incubation center I think so this is this is where the magic happens so the question that you asked is perfect and makes a lot of sense so let me just kind of go through this and I can share if you want to browse then you can share your window otherwise I can scroll maybe I can show it quickly so let me just do this yeah this is you now yeah perfect so basically deep kappa was born with this principle so the whole idea was to solve world hunger for AI meaning how can I make this as big as possible and out of deep kappa you know companies like URAI, real AI and also live AI which you just saw these projects are being born so the whole idea is to solutionize so you have to create solutions and research is at the core of everything we do so here you have your solutions and here we try to sort of bring you know we train our people and also we give workshops to customers and trainings internally as well and then obviously here this is where the magic happens where you're doing real-world projects with us so if you are a researcher engineer developer who is interested in machine learning I just opened up a slack to worldwide community just a few days ago so you can do that research is an important part of AI and I know we are too small I mean we cannot no way compare to these big companies like Google, Facebook or others but we do like to contribute and do our part here and so this was a healthcare project which we completed there's a new there are a few new researchers coming up so this is breast cancer neuro science we're doing a lot of stuff here and this is a paper and you have codes everything here so you can go to this website and sort of download code and test it so there's a new generated model which one of our research is built so this is totally new it's totally novel it's absolutely brand new you should try it and maybe you learn some things about it it's a flow based model for those who understand a little bit about the autoregressive models so it takes the best of the autoregressive and the flow based models actually that's what it's the intersection of these two I won't go into detail then obviously this is a paper I wrote together with with my neuroscientist researcher who's also head of research and a couple of others so this was another cool activation function researcher wrote that he's now working for a robotic company and that was also wonderful stuff he did together with a few other people there's also code available so that's research and this was my baby this is my personal initiative and I like to sort of believe that if you are not able to solve world's problem then actually you're not really sort of being very active we haven't been active on adding all this there's a lot of stuff which we've been doing since then so I go all over the world and I take sometimes risky bets going into places that may be difficult regions and I try to teach sort of the world about it so this yeah so deep kafa is basically just to sort of recoup back it's like an incubation ground where you can develop your product and solution with us if you're interested you'd need to go on deep kafa linkedin page company page and you can register and we're right now closing our cohort for this season we will start again in a few weeks and if you're innovative you have some really great ideas please do come over and take a look at it so very happy to sort of encourage people to do that so building is the best way to solve problems talking is fine I mean it's wonderful to hear you yeah I mean really it's great to connect back to you I look back at our old videos Nikolas all the long hair I had and it was real fun it was fun so thanks a lot I'm very happy to be here thank you so much for the future of the potential in Europe for something really awesome to all kinds of amazing new ideas absolutely maybe changing business as usual instead of going through you know all these issues with delay like sometimes the height is not getting realized fast enough but why not just make it happen so I think with smart AI and not just like it's going to be great absolutely I think the future is for those who want to look at technology from a progressive standpoint and use technology wisely, judiciously and making sure that we are able to I think because the benefits are huge and we should just continue to focus on the benefits solving the world disease, poverty social injustice using AI in medicine is huge so let's hope and let's keep doing that and for safe transportation self-driving cars but Google has been talking about these for 10 years but we want to see them we want this 500 billion to fund some self-driving car fleets why not have them right here and we have all the best cars electrify and put them out and we can we have the wind there's so much wind in Denmark they want to do huge islands but let's get all this stuff absolutely there's a lot to be done and I think we should do it we are trying to do it in our own way I hope in this experience we are also able to succeed because we are still a young company or young companies with Kourai Deep Kafa and Real AI and then other platforms that are coming let's hope that I'm trying to do my part I'm more than happy to that's why we are opening it up for people worldwide to come and contribute and work together also come to Europe and work with us so let's keep trying to bring a talent also let's distribute the talent as I said in the beginning in the most even way where each geography can benefit and grow and become sort of sustainable sort of technological power in their own region and still work collaboratively with every other region and then you will have all these discussions and frustrations and allegations they will all disappear because we'll be just negotiating and transacting and doing wonderful business with each other cool awesome so people can there will be all the links under the video follow you on the Twitter and follow you on LinkedIn and other stuff you're doing absolutely keep following us and very happy to sort of interact with you guys directly thank you so much Johnny Rory, Sravani you guys asked wonderful questions and hopefully we'll do some other sessions as well maybe we'll get some other guests I'll talk to Nikolas and who knows we'll do something again together thanks again guys see you take care bye bye