[Lecture] AI Super Powers: China, Silicon Valley and the New World Order

welcome to the Laconia schools public policy event book launch dr. Kai Foley's book on AI superpowers China Silicon Valley in the new world order dr. Kai Foley is the chairman and CEO of sinner vation ventures and former president of Google China he's got a very long CV so I'll just highlight a few salient points dr. Lee is president obscene adventures a artificial intelligence Institute sin evasion ventures and manages two billion dollars worth of dual currency investment funds and is one of the leading venture capital firms focusing on developing the next generation of Chinese high-tech companies right the foundings innovation into 2009 dr. Li was the president of Google China previously he held executive positions at Microsoft SGI and Apple dr. Lee received his bachelor's degree in computer science from Columbia University and a PhD from Carnegie Mellon University in the field of artificial intelligence taught early build one of the first game playing programs to defeat the world champion in 1988 as well as the world's first large vocabulary speaker independent continuous speech recognition system dr. Lee founded Microsoft Research China which was named as the hottest research lab by MIT Technology Review later renamed Microsoft Research Station this Institute trained the great majority of AI leaders in China including the chief technology officers and AI heads of I do Tencent Alibaba Lenovo way and higher while we Apple dr. leave that AI projects in speeds in natural language which have been featured on Good Morning America and many others dr. Li has authored 10 US patents and more than 100 journals dr. D has been a pioneer in AI research and has been in this business in the last 30 years so without much further ado let me call on dr. Li thank you very much sorry my voice is not cooperating today so I'm going to have a guest speaker give the next slide for me it's a great thing to build a better world with artificial intelligence the endo tchen of jones act IV idea so this slide shows you the power of deep learning because it is built on deep learning technologies and it could be synthesized speech by President Trump and that was not him speaking it was by the synthesis system and the deep learning is a big technology breakthrough that can do the following in a single domain keep in mind one domain only with a huge amount of training data reach superhuman performance so when trained Amazon clicks it learns how to make money the most money from people entering the site when trained on large amounts of President Trump speech it talks like him when trained on CT of lung cancer and non-cancer it learns to distinguish them from each other and when trained on go games it learns to play the game of go as superhuman capability so that has been the most amazing breakthrough in the past 62 years of AI history and in terms of application is actually over so many fields many people get confused and think only robots autonomous vehicles are AI but actually the same deep learning algorithms go through these four waves of artificial intelligence let me start from the first wave it's the internet wave naturally that's the domain in which we collect the most data and is automatically labeled we are guinea pigs labeling everyday for Facebook Amazon and Google when we use them and click it learns what you like and don't like and then it knows what to show you and that's how the websites are know better than you do what things you want to click on what things you want to buy and can maximize revenue what's more each of the implementation of deep learning for internet websites comes with an objective function Maximizer so Amazon can choose to maximize revenue or profit and Facebook can choose to maximize virality or minutes per user so imagine the job of a CEO suddenly got a lot of easier and that's also why these internet companies are near trillion-dollar companies is because they've had a lot of training data and this magical knob that helps them make more money now the second wave it's applicable equally to businesses that have lots of data to banks insurance companies hospitals government branches so on so forth anyone who's had that data repository can now use it to apply AI to it take as an example at the bank whenever I meet with a bank CEO there's always a poor person is sitting in the back of the room looking very grim that is the head of the data center because that person generates no profit no revenue is merely a cost center storing data for archival purposes but guess what that data center and that person has just turned into a mountain of gold because with all the customer transactions and history a company a bank can now optimize customer asset allocation can help make more money can help minimize default can help detect credit card fraud so that is phenomenal and similarly for insurance companies and so on but before you start thinking it is only to use buy to be used by traditional companies you let me show you an example where it can be used to disrupt companies we fund an AI loan application and it's an app that you download and when you download it you can fill in all the blanks such as your income your name your workplace and so on about ten fields but also you're asked to upload your phone data up to the loan application decision-maker and that data isn't everything on your phone obviously it's only things permitted by iOS and Android it's the same things that Facebook and Twitter take nothing more nothing less and what the system will do is decide whether to give you a loan based on what you entered and what you uploaded and so think about it if you were to walk outside the National University and run into 3,000 strangers can you pick 1,000 of them to whom to each you long 500 Singh and there goes 500 hours what percent of default rate when you get very high right maybe 80% 90% ok Singapore maybe 60% but still a high number so guess again what this default rate for this longer piece is 3% so how does it manage to do that because it has so much information from the phone that you regard as useless there's actually little bits of valuable information put together correlated with each other to make a very smart decision this obviously includes all the things you typed but also how fast you typed it that makes a difference because if you're a fraudster you're probably copying and type slower also has information like what they of the month is it is it before payday or after payday before payday is a good long after pay big not so good because you were just paid why do you need money it will also have information like what happens do you have do you have a lot of games you have a gambling app do you have your legal apps or perhaps you have all serious knowledge apps that makes a difference under law what model is your phone that makes the big difference very big difference and also what's your contact list how many people are in it are they real people is the person you call that really your dad so on and so forth these are all things that can be found merely using the data you upload and what's available on the Internet so just out of curiosity we asked how many such features are there there were 3,000 and just out of curiosity we asked what's the least important feature it turns out to be your battery level so why does that matter at all well if you're someone who has OCD who pledged to Foley on all the time you're probably a little bit correlated with someone who you turns loans if you're someone who keeps letting your phone run out of battery you're probably a little bit correlated with someone who defaults so obviously that is not an important factor that might be one billionth of all the information that's fed in but all of it is considered and in aggregate it's something no human can do moving on to the third wave that's when machines start to see and hear and sense with things like cameras microphones heat sensors movement sensors through the reconstruction devices even in the dark all of that gathers data that before was transient non-existent and now can be used to make smart decisions to give you an example you all know Josh vo right Jackie Jo he's got a new name in China he's the police mention you know why because in the last counselors he gave over 20 people were arrested from the most wanted list because the stadiums were were they were they had cameras installed and the cameras were connected to face recognition to the criminal database and as a result the police apprehended some number of people - some of them oops sorry your ID looks good my mistake please enjoy the concert but to the unfortunate 20-something people who were actual criminals and who thought they could for one night sneak out of wherever they were hiding enjoy a concert they were caught this talk is not about whether that's a great app or or one that worries people but it's about the power of face recognition no human can possibly even if you took the 1000 best policemen in China they couldn't possibly identify 25 criminals out of a field of 60,000 people in the concert because we simply don't have the memory or the face recognition capability and we get tired the fatigue overcome us can't possibly do that so the these hopefully the lone example and the example we're about the concert will show you that AI is far surpassing our capabilities if you satisfy the requirement one domain lots of data now let's in the third wave also we'll have autonomous stores that recognize your face your movement your gesture your intent that you pick up this bottle of water and you may be bothered by put in the basket and it charges you or maybe you look at it with disgust and put it back that indicates something turning an offline store with the same power as Amazon or online or maybe you drank from it then this put then the security comes out access don't drink from our water so all these things are detectable now in wave 3 and they will automate a lot of things with smart vision and hearing where for is is robotics and autonomous vehicles most people know a lot about those already but I can tell you robots actually difficult to do everything humans can do I'm on the board the Foxconn and I can tell you it's not going to be easy displacing people who make iPhones those that requires a level of dexterity and hand-eye coordination however there are many jobs that are stationary repetitive that can be displaced by robots for example watering plants selectively with just the right amount of water Eliezer based on the growth as observed by computer vision such as picking fruits such as doing dishes doing inspection on assembly so we're still talking about tens of millions of jobs but just not very high dexterity jobs an autonomous vehicle is of course the biggest of all pressures that will change the way we transport ourselves logistics delivery it will make life so much more efficient convenient it will make the air safer there will be full fewer cars in the road we will no longer have to buy cars and it will be a lot safer especially over time because one very important thing is more data makes a a possible and more and more and more day that makes they are better so the moment that autonomous vehicles are launched hopefully it's pretty safe and then five years later it becomes extremely safe five more years even safer so it keeps getting better and better so that's the fourth wave each of these waves represent something like five percent increment to save to the GDP also represent some five percent of jobs displaced this slide shows you the key things of what makes a I work massive data late very good labeling and the single domain and usually you need a lot of compute power and some AI experts so who has bi experts obviously America in more specifically America with some British and Canadians are the far by far the leaders in the world so why it was taught would I say that China has a chance because there's three very important observations to make the first is there have been very few breakthroughs so people generally assume there are a lot of breakthroughs because you read about them in the paper but actually all the breakthroughs that you read in the paper are built on deep learning or similar technologies and deep learning is fairly well understood and nor can we expect us to continuously come up with another big breakthrough because after all in the last sixty two years there was only one breakthrough why do we think there will be five more in the coming years secondly because the technologies are well understood were now moved beyond technology discovery and disruption were moved into taking the mature technology and applying them just like the early days of the internet the discovery of tcp/ip amazingly important the building of the web browser amazingly important the invention of electricity amazingly important but there were never was a tcp/ip 2.0 that disrupted everything there was a 2.0 but it was a small event there were never was electricity to point out that destroyed everything it was the one we know when the groundwork was done and then all the applications build on top of TCP IP of the browser of electricity so I like to say that deep learning is like these things so we're in the era of implementation thirdly AI is a very open domain all that is known is largely in the open source if you want to learn AI and take enough courses all the open sources out there you can build what you want to build so companies compete on their speed of implementation and the ideas and how quick they can make money not on the breakthroughs because the breakthroughs are done we understand them there in the open space now how does China build on these three things well first China has a lot of great AI researchers and engineers they're not as senior as the America once this chart you see of all AI papers 42 percent of Chinese off a forty percent two percent of all the authors are Chinese and Chinese can innovate in these products you see the Chinese used to be copycats but Chinese have become equal to the u.s. in the green stage and now has to have leapfrog to build three hundred billion dollars of value in the orange slice that you see and the Chinese entrepreneurs they are hungry they're good at finding business opportunities they work hard they build barriers because they're surrounded by copycats they have to build products that are uncouple the only uncopyable product is one that takes that is built a moat around it that's also hard to copy so for example matrons 600,000 delivery people and the infrastructure they built costing billions of dollars for example TD going into buying these vehicles leasing them insurance gas stations that locks up the domain so that's the Chinese method of competition it's a very good fit with AI they're built up over time their capital intensive they raise a lot of money and then they build the moats that's very hard to imitate the fourth reason is China has a lot of money a lot of money flow to China ai this is not government money this is private money and a lot of this money goes into funding Chinese AI companies Chinese AI funding exceeded us in the last year and as an example these are the five of our unicorns so these are AI companies we invested in that have become over 1 billion dollar in market capitalization and total value is 21 and the newest of these companies was founded only two years ago these are concepts that I think are are heard of or perhaps not even believed in Singapore but now you've heard it so you should believe it the fifth reason is the power of massive data the right chart shows you the more data the better it performs in fact in AI we have a saying called there's no data like more data anybody care to guess who said that a gentleman named dr. Bob Mercer the founder of Cambridge analytical a very famous esteemed AI researcher who turned into additional roles and in the MSO in the age of AI if data is the only and China is the new OPEC China has not only more people but also more usages Chinese people use takeouts more because in China you can get food delivered to you from five hundred restaurants in 30 minutes costing seventy US cents per delivery and that is the amazing thing that causes Chinese people to have more depth in usage and that's where the data comes from a lot of people in the West is soon a Chinese people just don't care about the company's trade data government has all the data that is not true the companies behave much as Western companies but this is that there are more people and they use the data more in particular I want to point out the use of mobile data is particularly important because mobile data is the most valuable data it means you're paying for something it's not just a click on the page but you're paying something it indicates you want something and that can be used as a rocket fuel to learn a lot of great data and finally Chinese government strongly supports a highly Chinese policies tend to be techno utilitarian which means you try it out and then regulate it only him issues occur so with mobile payment that may have been stopped in the US because credit cards may raise the issue that software companies can be hacked or it can be fraud or can be trusted with managing your money but China would trust Alibaba intention as long as they live up to their work and they were proven trustworthy so they've taken over at the credit card space and also China has an AI plan on the left side wanting to be the global best by 2030 and then with that plan each each enterprise in each city may come up with specific plan so for example the state-owned banks once the government said AI is important they might procure some AI software and city of Nanjing said our schools are very good let's build the world's largest AI Science Park and China decided to build a new city called show ad which has autonomous vehicle built in with top layer for pedestrians bottom layer for cars thereby avoiding the kind of accidents we saw in Phoenix with uber autonomous so as a result we anticipated that China will catch up with the u.s. somewhere between now in the next five years and that the most important to take is that China and us will be by far the co-leaders in AI who exactly will be ahead really depends on a lot of things this shows a projection China's slightly ahead but only in implementation u.s. is clearly ahead in research so with new technologies invented that could put us back in the lead but what is clear is that in this race there are not three metals like the Olympics there are only two metals and they belong to us in China who gets the gold remains to be seen but there is no bronze medal AI will create a huge amount of value about 16 trillion dollars net additional GBP but it will also bring a lot of challenges and due to the interest of time I'm just going to cover one issue which is job displacement that is with AI being able to do so many jobs are all our jobs when they be taken away well it is not if we think about what AI cannot do there are two sorts of things one is creative things and the other is things that require empathy compassion people-to-people connection so if these two attributes separate all the jobs and tasks that we do we'll find that in fact in the lower-left all the jobs will be taken by AI and that's of concern and we need to do something about that but the jobs on the lower right is a perfect example of human AI symbiosis with a actual helping scientists find more cure for cancer on the upper left we will find that AI can be the analytic core while the human provides the warmth for example in the case of a physician AI can do the diagnosis but the physician connects to the user to the patient here's gets the patient to tell all the problems and enters in an AI engine and provides the comfort and confidence thereby maximizing the likelihood of recuperation but also making the cost of health care much lower and then on the upper right side is where humans will excel in both compassionate as well as creative skill sets so we do have some things to worry about in the lower left or we also have a lot to celebrate on the other three projects but the most important thing I think is we look further out in the future I think your children for those of you who are students your children for those of you who are teachers your grandchildren they will probably enjoy an amazing life because by the time they look at the effect of AI they will only see that AI has liberated us from doing routine jobs allowing us to have a lot more free time to love the people we love to do the things we we're passionate about and to have time to think about what it means to be human and for those of you who are a little fearful of AI remember it is just a tool we're the only ones who have the free will we will control the AI tools and we get to write the ending to the AI story thank you I'm sure we have a lot of questions in the Q&A but meanwhile we will let dr. Li rest his voice the meanwhile I would like to call on one of the pioneers of AI in Singapore she works for the Provost office and previously we need National Research Foundation and she oversees she's one of the people who oversees AI research in Singapore can I call on dr. C that any time to describe to us what is happening in the AI feel in Singapore after hearing the global perspective of what's happening out there what you got so AI I thought it would be a good idea that you share what panini we didn't sing amor what what do we do in terms of like the I in Singapore and during this talk I will talk about our national programming a I caught the eye Singapore okay so what is the asking a more relaxing apart we can achieve it our mission statement over there anchor the international have oblivious blah blah blah all of the big words over there it's almost sounding like I'm having statement but then let me just try to translate this into what it actually means so it means that our government is actually forward-looking enough to realize recognize the importance and potential impact of BI on our everyday lives and then that we are then actively investing to build national of lead in the I to driven mainly from our ending comment in this case because at stage in Singapore most of our time who in a and she resides in the rnd community to prepare our workforce eventually to be AI ready and at the same time towards what and rip the potentially I as a nation okay so in hi Singapore is national platform is anchor that anyways but being national involves all the universities as well as the public research institution in Singapore and we also do have strong support from all our government and funding agencies okay so before I proceed for the describing of what they actually do I thought it's good to I should reiterate the point that actually our national investment AI is small it's really small compared to the rest of the world in particular our government actually pledged up to 150 million investment over years under AI Singapore so given the small local talent pool and it's also relatively small funding resource we have to be sufficiently differentiated from the rest of the world so actually how do we do this so how do we achieve the maximum impact given the constraints that we have so we think that we can actually do this by having clear objectives for and go in mind in things that we do maniac Singapore so this slide gives you the three strategies thrusts or like we love the quality pillars in the s and for big a national flat one there's we have to have something for everybody so let me just go through each of the killer one by one I'll start with the left understand so for AI research AI research is for the researchers this actually aims to it's about developing next generation AI techniques from the fundamental research in the pull off like camping next breakthroughs after the club you know we go and well whereas on the right hand side they are innovation our signature hundreds per month projects actually aims to help our companies to adopt you know to innovate and adopt their technologies so in the Middle Pillar we have FDI technology now my colleagues in India Singapore and NRF not this but I think this pillar the means the code of this pillar is actually our government and funding agencies so why do I supervise do I say that so this AI technology killer aims to demonstrate the technologist economic and social impact of AI by solving pertinent national issues in relatively shots well define period of time how we do this is actually by supercharging or cysts almost like the IRD program on steroids why because one we actually set a very clear goal objective problem statement that we are trying to solve on top of that we also specified a clear measurable impact that we want to achieve of the what we call grand challenge that we are trying to solve and suddenly we give our researchers access to high quality data what you saw the problem so it's really to put everything together make it as seamless as possible and help only suggest virtual address and start with a grand challenge that the nation is actually facing and in the hope that we can actually Dan demonstrate well achieve both advancement occupy at the same time also demonstrating the return somewhat of return of investment I should argue that what you do okay so I'll go through that each of the pls one by one so for AI research is about investing in research vital and strategic to Singapore so the question we ask are the things like what gaps in the currently I technologies what what how can we actually improve techniques to go beyond deep learning because we know right now deep learning although in actually is actually very mature and it can actually achieve we can stop and be useful in many applications but we all know that humanity is not learning the way that deep learning actually learns I mean human who does not need to actually look at 1 million images of books to actually know that isn't it is books and we also have everything that we generalize our concepts so there are actually much room for improvement for our research researchers to actually explore what is next we are virtually deep learning at this point is actually good to give some sense of the state of AI research in Singapore so overall we do have us small pool of high-quality researchers producing high-quality research output in fact we actually ranked first in terms of in the world list on citation impact of our R&D publications and in fact actually intelligence systems actually a eyes tend to watch tend to watch in 2018 four of them I actually based in Singapore the message okay this is actually the roadmap for RMD a Singapore we are a small player in terms of funding size in their research area in Iraq and that's what is actually therefore is actually important to invest selectively and we placed we should actually place strategic bets we feel that two areas that we can play bass in our first one is actually learning from small data set even the size of Singapore business a very relatively are usually small but hiring quality which that actually leverage that and one explain ability expendable AI because we feel that our government are actually very receptive of the ideas innovation and we could possibly actually in Nationwide adoption of VR technology and to capitalize on this we should best okay so the AI technology pillar aims to use I'll to tackle our national problems national challenges to AI Grand Challenges project the branch what I mean by Grand Challenge challenge the criteria that we actually set they are three verses that she has to be inspiring for the researchers is to be has to be interesting enough what we suggest what what you solve the problem he has to be impactful as if not that it wouldn't be actually Grand Challenge and finally it has to be measurable so the trick potential application domains that have been identified for the Grand Challenge our healthcare urban solutions and finance all of which capitalizes on our unique ability to get a large unit label data sets for to support a lifetime training so we have recently launched our first Grand Challenge sometime in June this year the grand challenge actually helps the statement is provided there how can we high health primary care teams to stop or slow disease progression and complication development in pH preach meaning that's behind the high cholesterol high glucose and high blood pressure and in by 20% in five years so that there are three are measurable goals or outcomes why we chose at least from statement is because it's actually taught three chronic diseases in Singapore it's also dropped three causes of Polytechnic attendance polyclinic attendance our top three also in eighteen to sixty nine years age group and we also estimated the project that the residents with three h2 which one point five million in 2020 so this is actually a timeline so well we have already we have just finalized the selection of the teams that are going to participate in the grand challenge and we are hoping to actually announce the announcement sometime in January so stay tuned then finally the last pillar is actually innovation in Atilla the mission is officially to accelerate industrial options by pinging businessman so under this track we have our flexible program we call hundra experiments where we I should be together attend the and searchers they are engineers to solve industry problems and at the same time train AI manpower for the companies and also the am is actually to come up with a Minimum Viable Product within nine to eighteen months for the companies based on their formal statements and data then finally for the companies we also to be also feel that there is a need to develop a particular range engineers so we do have talent development programs for industry becomes stuff from the bottom we have what we call the I apprentices are the parentheses program reality which actually aims to train AI engineers who are able to design develop and deploy their applications we just the number target number that we have us and then moving on weaved aim to Train to a two thousand engineers who actually understand and use can use AI actually appropriately and be able to achieve program basically I program and that's our a a I for industrial here for I and finally we are also we are sort of a program for AI for everyone is actually basically trying to bring a I to the masses to so that people are able to understand what is AI and able to identify what is consistent daily life lastly these are the list of our industry partners meaning those people that are having I've been in some sort of collaboration projects yes and let me throw the first question to dr. dr. Lee dr. D you mentioned in your book that you said that universal basic income is not the solution to these problems can you elaborate on that please sure universal basic income is the idea of just giving everybody money that's a very simplistic approach and first of all where do you get the money from for the United States who would cost three trillion dollars and who's gonna be taxes for that so for beginners you have to be selective in how you redistribute income but I think what is really missing more importantly is the fundamental issue that it's not just a loss of income it's a loss of meaning so when they I displaces jobs that are routine jobs the people have been attached to those jobs because that has defined the meaning of their lives and when you take away a job we have historically seen people with 12 months of unemployment even though there are so full social welfare have dramatically higher depression rates suicide rates substance abuse rates so whether we like it or not our generation and your generation are now many people believe that work is the meaning of their lives so there is no way you can solve the problem even if you have money to give to people and just say you're 35 years old you're no longer needed here's fifteen thousand dollars a year for the rest of your life that's going to lead to terrible problems I think the only solution given the way work is perceived in a society to is to responsibly retrain people so that they can be gainfully employed and find again the new meaning of their lives all right thank you we have two microphones to the left and to the right please keep your questions short and end to the point hi my name is minto I work for a venture capital firm as well I have a kind of a two-part question given that China presents a kind of it maybe it's a authoritarian perspective on how to govern people should China be in a ice for power especially given that humans imprint on the creation of a is so prominent right we've seen many studies on how when we create eiope it brings with it such as a racial or SiC sexist bias ease etc etc so should an authority an authoritarian government be in a a super power and the second part of that question is what does it really mean to be in AI superpower what are the actual repercussions what does it look like let's say when China becomes an AI superpower what does that mean for the rest of world is it means is it merely that it absorbs the economic impact just like say when the u.s. became a steam power economic or electricity superpower okay right so what are the repercussions when I say superpower China's superpower I'm talking about the companies in China just like when you say US isn't a superpower we're not talking about the US government as in a a superpower we're talking about Google Facebook and Amazon and in the case of China we're talking about the AR giants that are emerging both the internet giants like Tenzin Alibaba but also some of the unicorns were created so I don't want to connect the state with the companies we're talking to other companies and and I think there are a lot of assumptions that China as a state commandeers all the data therefore it can do whatever it wants that's just not the case I think you should go to China and you will see the Chinese companies operate much like American companies they have to be responsible in using the data otherwise the users will challenge or uninstall their application so I'm speaking as a commercial venture capitalists and my comments are about the companies so I don't think the I'm in a place there are any of us to ask theoretical questions that we have no impact over history go back to history you know every Industrial Revolution caused a global power shield so now we are you force in that's why we are watching it is just a part of the revolution so do you think China will overtake us in this revolution and okay briefly tell us the difference of the Indo business and Communist China and with capitals the US okay so if I answer China while we're taking the US government what say ah that's evidence we should escalate the trade war and if I answer no China won't overtake the US government the Chinese be covered and we'll say hey how could you say that so it's a no-win answer but nevertheless I think is really hard to tell I think Chinese innovation is different it's more about tenacious winner-take-all data driven saw Marcus eyes driven government supported type of building of basically very powerful companies that build modes that are very hard to cross and create companies that are very hard to copy that's the Chinese model the American model is more visionary changing the world keeping it very light and basically from copying so I think the answer to the question is which of the models are applicable in the future because I don't think China will get very good at the u.s. model nor will us become very good at the China model and that's why it's kind of hard to to answer thank you gentlemen hello my name is Philippe Emily's a young Indigenous a economies from Ecuador in South America so I have a question about the AI so how do you think the idea will impact in our economic and social topics in developing countries and the second question is what should we do to reduce the back in the economy and socio-cultural aspects thank you okay yeah I think the implication is that if there are only two winners out of the AI competition that means they take 90 percent of the profit or the market cap and then surely countries like you know EU UK Singapore and Israel Canada what takes a lot of the rest so that leaves very little for the developing countries and that is a big worry and in my book and as well as my recent article for blooper new economy forum I talked about it I think it's more important than ever to find the smartest people in developing countries and make sure they get the best education so they can come back and change the future of the country's secondly I think it's important for our country to figure out what's special about it that it won't just be dominated by us or China for example Indonesia the company called go check is a motorcycle delivery company completely different from uber because uber is very hard to get from anywhere from Jakarta in Jakarta to vintage ekkada you see the traffic jam is three hours wherever you go so the only way you can get anything anywhere is by a motorcycle that goes through the cars so uber doesn't know that Dede doesn't know that so to go Jack quickly grab the place and says I understand my country better so that kind of country specific innovation I think it's another possibility and but I think we have to also recognize another very bad news for developing countries which is that the China model the old China model and the old India model probably won't work for much longer maybe five more years because China and basically did manufacturing outsourcing for us for jobs u.s. didn't want China did it cheaply and then made money to crawl out of poverty India because of his english-speaking population that outsourcing basically call-center IT outsourcing and those were jobs Americans didn't want but they paid okay and that helped India crawl out of poverty but those jobs by the very nature that are outsourced opal to another country they're certainly outsourced about 2:00 a.m. and because they're largely routine those models won't work for developing countries so it's incredibly important that developing countries unless they're already like Vietnam already well on the way of taking men in fact away from building China then they can keep going for a number of years if if a country is nowhere in that scene I want to develop a China model or an Indian model from scratch I really caution they think very carefully in the future the only things that can potentially be outsourced is service so think about how to do service outsourcing not call center na by T and not main margin thank you dr. Lee going back to your point about the superpower rivalry where what what advice can you give to small players like Singapore who would be left with the crumbs well Singapore has a number of advantages it is one of the very few countries that can attract an average IQ higher disproportionate from its population right us has that power Singapore is that power because it's a magnet you have immigration policy you're welcome immigrants so smart people come here to study or to work and many stay some go back many stay so how to leverage that higher IQ I think it's going to be very key that's an advantage Singapore has fantastic financial community Temasek as well as the markets are excellent and I think Singapore occupies a unique place in Southeast Asia which is a large market despite Singapore being a small one it's a messy large market because it's not same language same government same culture in fact there are substantial disagreements but nevertheless it is a region that is sometimes able to work together so it's possible and Singapore has a strong China connection because the largest Chinese population so I think that he is how to use these four features built on what was presented earlier on the Singapore plan and and I think make a bolder step what you've done is great in the foundation but the question is can you make a bolder step of saying we're going to make that bronze you never and here are a lot of things we're going to do because what you've done is very good I think all the basics are done there are no errors many states you've covered all the things they're recovered but can there be one moonshot that will propel Singapore to that bronze medal that would be the question in other words is there another joseph schooling for a tie-in signal I mean Singapore being small we also we we need to select selectively blessing so I mean all the monies that are foreign if I both by public and private sector how do you think the public sector embodies academia are indeed should actually differentiate ourselves from the industry so that we are not we can compete with an industry in terms of our RIT so how does should we actually position ourselves to support the industry and - okay I'll give a two-part answer one on the ecosystem side one on the public sector research investment side so first and in your maintenance humane question when you want to support great research in Singapore there are sort of two paths one is to encourage great research that is just great that is the next deep learning the other is to practice a lot of people I saw both aspects and obviously you need to do some of each but this but I would say the second one is a lot more scalable and and I think any country can still put out a claim on that as an example in my company we train 300 students in five weeks at the end of which they were building amazing AI demos and systems that actually started - some began to be deployed in real scenarios so the speed to train so I think if you look everything yeah I think everything should be done with a long visibility to the future and one thing that is clear is the barrier of entry to AI is getting lower and lower and lower so given that if you want to follow the trend how do you make the numbers you have even ten times bigger and how can that set of people can think of or attract trained retain these people wherever they are imagine the training we have for 300 people could probably be done for 20,000 people if you created the world's largest training camp in AMI and then offer the top graduates a citizenship maybe that's too much ideas that's a good idea or if you another idea is go to America look for all the AI PhDs we can't get decent give them all a citizen sure I think some I think about that's what I mean by moonshot that is big and the other point I wanted to add that you didn't ask about and anyway is that you know you all know firstly right the yeah Corina so you need more people like him I think he came to Singapore kind of serendipitously but you need to go and grab these people to Singapore because the entrepreneurial community organically isn't easily going to produce five or ten for sleaze and grabbing them similarly you need to grab like five or six really good feces don't go for numbers go for the super top ones look what happened in China right in the very early phases it was Neal shed and Hugo shone who invested in Ponemah these were what sow the seeds of Chinese entrepreneurialism this is brilliant amazing theses helping brilliant entrepreneurs grow who helped intern for the VCS to grow so I mean you have to Massey you've got the late stage covered but the early stage you got to start from the beginning yeah so this C okay thank you gentlemen my name is Daniel Wagner I come from New York and I'm also at the author of a new book on the AI as called AI supremacy which is all about the race between companies and countries I agree with you this is a race between two countries largely the China and the US and my question is there is a certain amount of responsibility attached being a leader in this arena and I'm wondering what your views are on the Chinese government's responsibilities and company's responsibility to develop AI responsibly and specifically visa be the development of China's social credit system I think every country and every company should have the same responsibilities as well whatever country whatever companies they're in and I think the Chinese companies are beginning to recognize that responsibility that's why I do entered partnership for AMI and that's why I'm participating as the co-chair of the World Economic Forum's Council for AI so that is happening and I also see Chinese think tanks and I understand NUS is working with Chinua on some of the projects about AI impact on society so these things will happen you know with great power comes great responsibility so I'm fully confident that will come to pass Social Credit System I know very little about it other than what I read in the media I have a session with a lot of Chinese people and I asked them how many of you know what it is nobody knows so until this rolled out I think it's hard to it's probably unfair to assume the worst but once it was rolled out I'll be happy to discuss it I don't know what my rating is I don't know if there is a rating I don't know if it's for credit only or for something else based on what I could gather it's largely for you know getting loans and approvals equivalent to the function of something like an Equifax in the u.s. it just happens to be state-run and not privately run but it may be more than that and the documents say more but we'll see implement their I hate to speculate something that may or may not be implemented okay for now it's a fake used I think it's real but we just don't know what it is it's not implemented yet okay thank you my name is Jaime NUS researcher so my question is regarding AI education because I'm graduated from the social science background sometimes I feel really stress because I know technically I don't know how to contribute to AI and then the AI train will replace some of the jobs so I want to know do you have any tips that I can people like me can identity ourselves and engage us in the AI trend going beyond just being users or even database thank you sure read my book I'm actually serious because I think it is one of the few books that can really be understood by and anyone without technical background once you read it I think gives you the basis to see how far you want to pursue I think it's important that everybody has an understanding of what AI can do the benefits and the dangers I don't think everyone has to go into the details which is why I worry about the many books that have been written when they're written by AI non-experts they often have mistakes in them when they're written by experts they're often hard to comprehend so one of the reasons for writing the book is so that is to write an accessible book that will give you a good introduction so please do that absolutely to follow up on this the product question about the implications for our education model our education model was built on it employment basis so what were your thoughts about the future of education in this world of they are in these disruptive technologies well I think young people would have to focus much more on things that a young cannot do and that is creativity and the ability to learn stretch strategic thinking and also compassion communication empathy so these are the foundations of what's most important any rote learning that you have done will definitely be better done by AI it doesn't mean you don't want to do any rote learning because you got to learn something memorize something before you can start to create so I think if we just keep in mind that AI is a good tool that it can help us deal with quantitative routine tasks and that we should do things beyond that including in education for example we are big investors in not just AI but education we invest in many education companies that try to let teachers to what humans do well and then outsource what AI and technologies can do for example one teacher can teach a class of 1,000 interact with a clicker for example exams can are better given on a personalized targeted basis by a machines and graded by machines same with homeworks for example a is better at being your English pronunciation corrector than humans can afford it and a is better at drilling math problems where there might be in inadequacies so when we offload these from the teachers the teachers can do what teachers do best that is being a person to a person mentor a very personal trusted friend and teacher who can help the students sir you know personal issues and give personal advice so I think that's kind of the direction 1c doclet cents for your lecture I'm a PhD student in the mechanical engineering my research focused on the 3d printing and the biofabrication is more relevant to the experiment so my question is that how do we improve our research if if our research is not in the order Diplomate is mechanical engineering yes it's absolutely related all the future of robotics automation factories manufacturing it's it is the big issue challenge a bottleneck it is why the robotics don't work as well as they do so we very much rely on you I think many researchers there are research area focus on the for example of the biological experiments all the material synthesis so it's more relevant to is more experiment based so what for the researchers in this area how do they improve their research to be a better researcher well every scientist can use AI as a one AI is good for it's good for data analysis good for filtering ideas good for trying numerically all the possibilities trying to research and and analyze what's out there and the human is good for setting hypothesis improving them and their intuition and their knowledge and wisdom and collaboration across cross domains and translating from other domains so there's a natural synergy I'm Andy from the national volunteers of the trophy and part of the work that I do Envy PC is to build an open intelligent platform to allow people to come in you know to key to contribute to Singapore in many different ways and then by doing so we then collect actually all these data and help every individual in Singapore ultimately to be able to build on a future so this whole aspect of AI and data it's very relevant to what we do and the reason why I'm doing this is that because you know I agree with you actually in the coming years we will have greater means to leave but not the meaning to default and you know would be smarter but not more human so true giving we hope to then create a caring and inclusive society you know for our future my question then is that what I'm doing it's really love what it is actually for a good intention but I will still be in a way using technology to share a menu like people towards actually hopefully they not to build a better person I would like to see your views on that you know especially relating integrate into the as well ethics because we now start taking orders by your data in Ferreira and actually combine them together and Shelley pop even if it's for a good reason is that something that in some way is ethics so can you repeat the question precisely yeah the question is that you know you don't work that I'm doing I'm actually trying to shake people for the future which is actually towards building and caring and inclusive society for Singapore you know so it doesn't discuss the court action aspect of shady people is actually many bleeding while shaping people is a very dangerous thing that's what Cambridge analytical did so I think you know I understand your Institute is doing a lot of research you should work for with NUS how things like that can be done you should study how people have done things like marketing and influencing people to buy a product and I think that's the same kind of technology that might be extrapolated to to teaching people to be more inclusive right I mean AI basically is just a tool you need to give it some objective function to satisfy so what you've described is a little bit too high level I think you need to get down to you know how do you measure someone's so-called inclusive activity and what are your actionable things that can change that so you can't just talk about AI on the grand level you have to have specific content steps to execute based on people's measurable things then you can connect them together okay we have time for one more question before we wrap it up please hi dr. Lee it's very interesting to see like you mentioned different models for example the u.s. model china model and indian or even though unique Singapore model and you mentioned like us model is more research driven and China model is more like they'd have driven accurate application driven and so in the era of rising of the AI and I wish to our video before as you mentioned like it would be an ideal case in all these countries can join together to cross the country boundaries and but how like they can come together to cross the such a boundary and and what do you think like human can benefit from these kind of ideal world cooperation okay I think we need to more than ever come together but the the forces of nature seems to be moving in a different direction and why do we need to come together because the strengths are different I think China knows how to build Chinese companies know how to build a very strong winner-take-all centralized system that's wins against competition that's a very knowledgeable thing and tested in the Chinese market Americans know how to innovate and come up with new ideas no one has ever thought of before so in a hypothetical world if I were an investor and I could get 50 percent US money 50 percent Chinese money 50 percent US founders 50 percent Chinese founders we can both throw the most immiscible company that can make a lot of money for human good right also there are a lot of different practices that might be very valuable as we face these challenges brought about by AI we talked earlier about developing countries how to have a gifted and talent education and perhaps Korea is a country that has done the best in the space and we talked about new kinds of jobs of service jobs perhaps Japan has done the best in this space perhaps there will need to be craftsman jobs maybe Switzerland is that the best there will need to be volunteers and volunteerism can be good and you know Canada may have done the best in that space so there's a lot of room for sharing it's just that is both a vein how to build a great company level and how to deal with problems of AI level and and there needs to be forums for sharing given the countries are less likely to work together in the future that they have in the past so I think it's incumbent on us as the private actors and the private sectors to try to do that now that the countries seem to go off in their own paths fortunately many questions very few time please go and by dr. Lee's book Kinokuniya it's a bit also available in Amazon and you can watch him on TED talk I'm sure there will be more opportunities when we can hear from dr. Lee and in your next visit so please join me thinking doctor [Applause] [Music]

Channel: Lee Kuan Yew School of Public Policy

Views: 12,288

Rating: 4.7611942 out of 5

Keywords: AI, artificial intelligence, superpower, US, China, u.s.a, america, digital disruption, silicon valley, new world order, kai-fu lee, lee kai fu, Sintia Teddy-Ang, public policy, lkyspp, lee kuan yew school of public policy, lky school, nus, lecture, ai superpowers china silicon valley and the new world order, technology, machine learning

Id: gsFFipyt6Ns

Length: 70min 15sec (4215 seconds)

Published: Sun Jan 20 2019