 I think we're ready to start. Good morning, good afternoon, good evening everybody. My name is Olaf Groot. I am a professor of practice at UC Berkeley's high school of business teaching technology and strategy courses. I'm also the CEO of advisory think tank Cambrian Futures. Welcome to the session which is live streamed around the world and I want to explicitly welcome not just you all here in the room but also our audience online wherever they may be to this very exciting discussion about a cutting edge if not the most cutting edge conversation that is going on around the world right now on AI readiness amidst an AI revolution a cognitive revolution as it were. We have a very high caliber panel here assembled today and let me just say a few reminders. This meeting is not subject to Chatham House rules because it is public so you can quote and you can use what you hear for your own benefit and that of citizens around the world hopefully. The other thing that I should say is that if you wish to tag or post this event please tag the world economic forum at hashtag AM and C for annual meeting of the new champions 23 AM and C 23 that would help all of us spread the wisdom of the panel here today. Before I introduce the panel let me just frame up the topic by saying that AI and especially generative AI has been touted to bring with it the potential for very very significant economic development potential growth to the tune of some 15.7 16 trillion dollars around the world over the next decade or so. These numbers on generative AI are something like four point four trillion likely added on top of the 16. Now all of that mind you of course is theory until we make it happen. So in order to make that theory happen we have to build ecosystems. We have to create conditions that allow us to exercise the tremendous potential of this new foundational technology. Is there hype most definitely there is hype that does not mean that this technology is not groundbreaking and won't change both our economies and our society. So we are here today to discuss the potential for both the economy but also and most importantly for humans to put humans at the center of that development. So with all that said let me just introduce our panel briefly. We have with us today her Excellency Paula Ingebirre the Minister of Information Communication Technology and Innovation of Rwanda. Welcome Minister. It's a pleasure to have you here with us today. We have Leo Jiren who is the chairman of a very successful AI driven company New Soft Corporation which is a mature company employing AI here in China. I was told 40 percent penetration in social security 30 percent penetration in hospital AI systems or digital systems. So a company that is already very successful spreading AI. We have also Darko Matowski who is a startup entrepreneur very important part of our ecosystems of course Chief Executive Officer of Kaoza Lens which makes explain AI explainable as it were very important for trust assurance. Trust assurance will be a key theme here as that is the most important currency in what we have called the cognitive era. And then we have Joanna Bryson my colleague at the Herty School not my colleague but my my peer at the Herty School where she's a professor of ethics and technology. She was also in the first batch of experts at the global partnership on AI. So has been exposed to the topic of development of AI that is ethical and trustworthy. So welcome all of you please let's hear a round of applause for our panel. Without further ado I'd like to start the conversation with a round of opening statements please from each one of you and I'm going to turn to Her Excellency the Minister and Rwanda and I'm going to ask you Minister how do you what's the agenda of tech development in Rwanda and what role does government play as well as what's the interface between government and business and other stakeholders to make those new frontiers happen in your country. Thank you very much Olaf perhaps I should start by saying Rwanda sets itself to be a leading technology hub on the African continent and to do that our strategy is to be a proof of concept hub and one would ask what does that mean really it's you know being a space where innovators startups big corporations can come and experiment with emerging technologies test them try them out and if they're proven successful then they're able to scale to the rest of the continent from Rwanda and that has been a very deliberate strategy because I think for a very long time as we thought of our ambitions of being a leading tech hub I think one thing stood out as a major challenge which was always the size of our population about 14 million population and for many big companies that's felt like a too smaller market to expand into and so we figured if we're really going to nurture new innovations and technologies and really be at the forefront we might as well position ourselves as a proof of concept hub a market where really you don't need a big market to test new technologies I think we've had very much about zipline and how it started off in Rwanda using the drone technology to deliver emergency health products including starting with blood products and so once that concept was proven successful they've been able to expand in about five other markets and so really that's our vision as a country and very particularly as we talk about AI I think let me start off by saying the role of governments is to put in place a regulatory environment that enables innovation just about a month ago we did put in place our national AI policy which we put together with the support of the Center for Fourth Industrial Revolution in Rwanda and obviously with the World Economic Forum and for us it was very critical that as we think about it Fisher Intelligence we're already putting in place that enabling environment both for business regulations and policies that is going to allow with the experimentation of AI solutions and I know the other panelists will talk about you know AI in healthcare I think that's one particular area that we're looking at to see what healthcare solutions we can you know AI enabled healthcare solutions that can be deployed we've already done you know an economic analysis assessment to see what are those industries and segments where we feel like AI has the most potential to disrupt but also create impact and so healthcare agriculture public service of the three top ones that we've already identified with very specific use cases that we will be implementing over the next two years to really look at what the potential of you know implementing these AI solutions will look like and then maybe finally your other part of the question which was really around what is the you know the interface between government and businesses I think as I said one is the bit where we put in place the enabling you know regulatory environment for innovation to the and what does that environment look like it's really how do we enable access to capital and the state financing for these startups and businesses that need to experiment with this how do we create a market for those innovations knowing that government is the biggest consumer of these technologies I think with the hype but also the fears and concerns around AI to build trust government leads to needs to lead the way in implementing some of these AI enabled solutions and creating that trust that is required within the population so that also the private sector can be able to do that the other bit is really opening up data I think data is really the heart of all these AI you know models that we are building and government has lots of data available so how do we clean up the data structure it beta and how do we anonymize it so that we open it up for you know the industry so that they can really leverage it and create you know breakthrough innovations. Thank you very much Minister you already said a number of the magic words here of course including the word trust because as you build an ecosystem in the country you don't just want to enable entrepreneurs that's key of course that's core it's core to economic development it's core to this AMNC here as well but you also want to make sure that these innovations fall on fertile ground and that they do right by the people of Rwanda and so coming to the the topic of trust and of course trust comes with understanding it comes with understanding it comes with a certain level of skill of literacy as it were of not just basic digital tools but also of artificial intelligence what it does how it does things how you know you deal with data data privacy data agency and things of that nature and maybe we can come back around on this question I'd like to next then go to to Joanna on on this to ask you know this this whole topic of trust how do we look at that between business and government clearly I don't think I'm saying anything new when I'm mentioning that we have not been doing really well on the trust front when it comes to you know big digital platforms when it comes to data when it comes to artificial intelligence the feeling the prevailing feeling has been that we are objectifying too many people around the world how do we fix that how do we make sure that countries like Rwanda China any country around the world fosters AI readiness in a responsible way well thanks very much for the question I just want to say quickly as someone who is now a professor of ethics and technology just to be clear about this I also was a programmer for five years and my PhD was in AI from MIT working on transparency so we actually have met before so so when I say this I I'm not saying it because I'm a technophobe or whatever I really like the way you framed the question a lot of people especially the British used to ask me how can we make people trust AI and that is the wrong question to ask we need to be talking about the people who are building AI and the people who are regulating AI so I love the question you actually asked which is how do we make people feel secure about this I think it's by making people secure and we need to be thinking very much about the fact you know are we worrying about wages how are we helping the people who are displaced when people have an unexpected life event are they likely to be able to continue paying their rent or their mortgage right so so governments have an important role in helping us all deal with change because we're entering change whether or not with technology or without having said all that I do think part of the reason that we have challenges with trust right now is because we know we're not getting the information we should and we haven't had enough transparency so I do think it's really important that we recognize that we're in the information age that we find out a lot of things a lot of interesting ways and we do have to basically do things that benefit our people and or else the people will find out and and sometimes governments find out from unfortunate voting events but you know we need I think the that we should think of AI hopefully and not only is something that empowers us and makes us all more productive and can help our economies but also as helping us to be able to understand that the governments are doing things for us and that we can see the value do you know during the pandemic populism actually dropped in every country except the US for some reason but by and large people could see that their governments were doing something and they hadn't sort of understood why pay taxes all of a sudden oh there's a big problem the government is helping me so I think that we need to think of the age of AI as an opportunity it's an opportunity for knowledge and for a currency and for transparency not only with the technology but throughout culture thank you for that I'm going to ask you a follow-on question because what we heard from the minister is of course that you need hubs for entrepreneurs you need to enable entrepreneurs that's where change happens in a society that's what upgrading of a societal skill base happens where the next tier comes to fruition of economic and social development you need infrastructure for that you also need to make sure as you said minister that you flow the benefits of that through health care systems education systems you mentioned a number of domains that really have public benefit what i'm hearing from you of course Joanna is also that as we do that we have to make sure we keep the human at the center right yes and and so how do you so so how do what's the element here that we need to bring to fruition governance is it you know what do we do to make an entrepreneurial system like this not just successful economically but trusted by the people well like I said I think well there's there's two problems of course one is opening a window in such a way that people can understand it and that doesn't mean you have to open source everything or whatever you know open source and people see all your code we can still have competitive secrets and advantages but we need to at least know what there is to know and what you need to do to be able to find out about that so for example in the EU we've been talking a lot about audits and some companies are very afraid of that but you need to understand other industries not not the the vast majority of industry now is all digitizing and they're not afraid of AI audits because they already do compliance it's just the tech industry hasn't been used to compliance yet so I it the fact that you will be audited and that you need to be able to yourself know how your code works I mean some AI companies have been very sloppy means that it shouldn't terrify you and it does not mean all your competitors are going to come through the door and see what you're doing and that you have to explain every detail like every weight on the neural network no we audit banks without finding out how the synapses work in the heads of the people the same way we will audit whether you whether you actually followed good practice best practice just like any other industry do diligence best practice whether your people did that when they trained the AI right and if they use machine learning on their AI then when they trained it when they tested it how are they ensuring quality so that's the kind of thing that companies can do and that now the EU looks very likely to be mandated in the companies do especially where AI takes decisions that affects people's lives like welfare systems or banking or education or medicine so so clearly the point you're making is that in the cognitive era with AI ethics and economics are tied by the hip we have to co-think that in order to enable good economic growth you're going to have to enable ethical growth I said there were two things I try to get say the second one the second one is of course political polarization and if you can't keep your if you're if you have a bunch of that the society that identifies as hating you then they aren't going to trust you so that is a that is a separate question I don't think we're touching on here but it is something that needs to be acknowledged we might come back around to that I'd like to next go to Leo Jiren who as I said leads new soft and we mentioned that it takes all kinds of parts in an ecosystem we need government enablement of infrastructure of talent of capital you mentioned of ethical safeguards in order to empower that growth I think it'd be a mistake to think of you know governance automatically as a limiting factor of growth but you have lived this life as it were as a very successful business leader and and so I'd love to hear from you what does it take from your perspective in terms of business government collaboration ecosystem building for you to be successful and to spread AI in a responsible and commercially successful way and you mentioned to me as we were preparing that especially in healthcare which is one of the big things that your company is focused on orchestrating that ecosystem is really key and there needs to be active orchestration over to you okay as a technology company we very much like to talk AI because AI is a fashion AI can create value but if you're talking how to create a value to social development how to convert those kind of technology to empower other industry I think it's a big very big challenge the challenge is not technology only it's an ecosystem I just take the example about the healthcare if we're talking about the healthcare transformation they adapt AI technology the firstly you meet a kind of a challenge about data how to capture data how to collect data privacy the governance and also if you use that data to make a diagnostic that is not like a charge ETP charge ETP is okay you can write an article write or not but if you make a decision diagnostic you must very precise and now any others choice should be at least same as doctor's level so I think that now the people are talking about charge ETP it's it's a booming it's happier because it can draw the picture you can write article you can do anything it's a kind of entertainment but if you're coming to healthcare it's a very serious and you need to talking about who paid money for this AI what is secure and safety how to convert the knowledge of doctor become a digital clinical pathway so but if you look at the demand it's a really huge demand especially in China China is a country we have a big population the most important is the challenge of healthcare system is big city have first class healthcare services second year city is second tier the services the countryside is another one the gap the quality of healthcare services is big the big cap means the cost is difference low quality of healthcare service means cost more money on the other hand China become aging population society we already have 200 million population more than 65 years old and years old so how to access the healthcare services equally that is a big big challenge so that is a market we can apply the AI so that is our practice we make a callot hospital we connect all the hospital in city a doctor nurse like him to deliver their services come to home at the same time we digitalize all the clinical pathway from individual doctor to be digital plan for so we focus on each of these cases we use AI to create a kind of dedicated treatment of digital you know the way to like a doctor in big city as small city or countryside they have the same kind of quality so it's very very hard because we need as a company we need to talk so many people we talk with the government officer we talk him that's good for people good for the city then we talk about with the hospital to say health hospital already make a big transformation from inside hospital to be home you are bad not only in hospital in outside of hospital then we need to wake up all the doctor is use a digital way AI so they take a very long time very very challenge so that means today maybe we need a good technology but I think technology is much behind much at the one thing than our ecosystems the most of challenge to applying the technology is we are not ready I mean the ecosystem is not ready to use AI today we can say every people use AI is for our daily life for personal assistance if we use vertical for industry it's still need cooperation cross industry or public and private sector have a more close cooperation I think it behooves us to point out that at the point you're making one of the points you're making is look this is not like chat gpt going to one million users in five days we're talking about sophisticated tools that have severe hopefully positive impact on patients but it might not if we don't orchestrate correctly you're also saying that AI can help you leapfrog over missing medical infrastructure and health infrastructure in tier two cities and rural environments that's a big so where where government required here what what do they need to do to help you orchestrate that ecosystem because I'm sure there's only so much you yourself can do how do you get doctors to accept this and do this how do you get locals China is very digitized of course but I suspect you still need to have basic digital skills infrastructure in order to make that leapfrogging work we we become not only technology we become a kind of educator you know we need to donk a door to government we're talking with him what it means color hospital I came to know that is a perfect selection to to the cities and because it's a benefit to everybody in cities especially for elderly person they are very hard you know the China single children maybe not living together with the family and also so many senior people living alone so how to like him to come to hospital we say color hospital is hospital at home so government is happy for that program and then we like him to to call China is good because most most hospitals own by government they like him to come to plan form to say you need to not only serve the people come to hospital you need to serve people at home because they work together with the pair side because insurance is pay at home is much less cost than pay at a hospital and then we digitalize all that kind of services the path clinical pathway for example the people stay three days in a hospital when searching after that they may take three months for recovery at home so they need to occupy the bed in a hospital so the hospital is very happy for those kind of turnover because they can improve their you know their margin for their services at the same time this plan form offer the big job opportunity we use this technology in one of our province in Zhejiang province it's now already it's around 10 cities that our systems thousands the nurse and doctor they come to the home of a patient and but we take a very long time you know it must to talk in the first they they don't believe they reject and the most important the non-regulation and the government is starting to make a regulation and how to use that data and how to like a payment and also some kind of standard for each of the cities it's a very very detailed so detailed just like a build real hospital but that is not the anxiety of the past it's very you know initiative so what I'm what I'm hearing then is that yes it starts with a government mandate because it's in the interest of public health but it takes a very thoughtful specific design government needs to work with the systems providers like yourself but also with the doctors because I'm sure doctors have something to say about you know providing services at lower repayment fees adding those services for rural patients on top of physical patients on location the patients will have something to say you have to make sure that they yes they're digitally connected but they also need physical infrastructure at home etc so it's an ecosystem orchestration a problem and I want to go now and take that to to you Darko you at Causa lens are are helping us understand better what AI can do understand help doctors who are not maybe digitally native or AI native help patients who may be even less so understand what AI actually does so I'd love for you to tell us exactly what that's about and I also then want you to answer the question of from an entrepreneur startup entrepreneurs perspective what parts of the ecosystem are particularly important to you yeah absolutely I will start by saying first that 85 percent of AI projects never leave the lab and there is a fundamental reason for that the fundamental reason is that people don't actually trust the algorithms the way it works is there is a usually in a bunch of data scientists they throw a lot of data in a black box and something comes out and as you mentioned you know it can be very entertaining if it's you know in the context of generative AI and you know generative AI has lots of uses and it's great but when it comes to decision making people really need to understand what the algorithm is doing the AI must explain why I made this decision it must be able to explain what it will do if a data point that it has never seen in the past comes comes to life I mean that that's the real life we can we have to have AI that guarantees outcomes even if we haven't seen a data point in the past and that's why there is kind of a fundamental research in AI and there's a lot of new technologies coming out that are able to explain why I make this decision are able to explain their decision even beyond kind of historical data on which they're trained so until we we solve this problem of trust we will have 85 percent of the project whether it's in healthcare whether it's in government whether it's in in any industry they'll just remain in the lab now luckily people like us and a bunch of other entrepreneurs even large companies like like Microsoft and Salesforce are working on emerging types of AI that can answer why and so the technology one of the technologies I want to mention it's called causal AI that is becoming a kind of go-to technology for explainable AI and trustworthy AI and actually the year regulation explicitly mentions understanding cause and effect and understanding the why as a way to to guarantee that that these models can be deployed in the real world and this is really really important for decisions that impact well-being of humans so kind of high risk applications in those instances we really need to understand the why we really need to understand why the AI made that decision and we really need to understand that this AI will guarantee outcomes even if if they haven't been seen in the historical data because the real world is you know it's it's not you know it throws a new data point that has never been seen in the past so we've seen that with the pandemic where a lot of models actually stopped working because it was a new environment that it was never seen before so clearly understanding is the first step to learning to influencing to shaping and AI readiness has to of course enable people to shape and to use in the right hopefully responsible ways how are you as an entrepreneur looking at government though to help facilitate that clearly it's a foundational issue a foundational thing that needs to happen you as an innovator as a small startup innovator do you have time to deal with government to deal with all kinds of stakeholders to orchestrate them what is your expectation yeah I think the European Union took a very very you know direct and strong view on regulating artificial intelligence they were fast to the party and they've just you know had a proposal for how AI should regulate I think there's a lot of good in that legislation there's a nice tiering of what is a high-risk critical you know no risk at all where it's not regulated so things like chatbots and you know things we can have entertainment with are not regulated and I think that's the right way forward I think the EU is has taken a really good approach to to the problem I think the reality though is that it's really you know it sounds good on paper the question is how will it be implemented in the real world you can have a regulation that is you know has the best intentions but if it's implemented incorrectly or enforced in the wrong way it can hurt creativity it can hurt this emerging technologies that are actually going to help us go from almost no projects in in the real world to most projects in the reals I think that's where we would like to focus the attention and I think it's about the implementation of those regulations because I think all governments have a good kind of intention at the beginning and it's the right thing to do to regulate such a powerful technology but the implementation has to be in the right way yeah so protection good but it needs to be appropriate it needs to be practical for the entrepreneur otherwise innovation is already dead in the water right I want to throw the gates open on on essentially two questions and and you pick if you like right so the first one is we heard orchestration who is going to orchestrate we of course default to government government is necessary we also hear frequently government doesn't have the skills to to orchestrate on advanced digital technologies not everybody is advanced as Rwanda or China on this matter right so that's number one number two is really so so on this matter of of trust where do we take this from here how for instance do you you institute trust in that ecosystem what do you do to get doctors to to play along and we can answer that from all of our vantage points right maybe the third provocation is you know let's think in a national clearly you have benefited from and the international community has benefited from your experience you have benefited from the international tie-ins right but it's easy to say this it's a shining example Rwanda is but we also know that internationally it's a bit of a mess we're not seeing eye-to-eye on governance on regulation on how we innovate right so this international perspective where do we take all that because you know innovators like like you and you of course Darko need to scale out globally eventually right to make the business case work so pick if you like maybe we'll start with you Joanna oh that's great I like it because I have so many things to choose between uh wow I guess if you're talking about who is going to do the orchestration I so I don't know if you can tell from my accent I grew up in America I spent 20 years living in the UK and now I'm living in Germany and I've been working a lot with the EU on how to regulate a lot of my former well I'm still American a lot of people from America are terrified of EU regulation doing the wrong thing or somehow shutting them down or facilitating competitors globally by putting extra weight on only the good guys you know they and again this is a problem you're seeing mostly in the digital sector because they were so under-regulated for so long they're relatively new and people tend to be hands-off about new things when I go to meetings where it's you know like for example the Euro Chatbot okay you might think that sounds like a lot of AI companies but it's mostly things like banks and healthcare systems that have to have natural language interfaces those guys say oh I spent two weeks a year on compliance it actually helps me the EU tends to to steer me to looking at things that help my cover my company and then I spent half a day on the digital compliance it's nothing what is it with these digital companies they're a bunch of winers right so that's that's what I'm hearing from from other sectors that aren't digital or aren't purely digital so I think it's important to realize that the sectors themselves are a lot of the people that are helping with this orchestration they normally what they do is they bring their concerns to the government and say help us coordinate we do national strategy in the EU we also have this coordinated strategy so that we can talk to really large tech companies and I think that's the other problem at this point we have a number of resources that we're using almost worldwide and some some regions have their own but you know email systems social media systems they're really transnational and they have a lot of power and we've been allowing them to do a lot of the orchestration and they have not been transparent they haven't even worried about things about like where is their data stored and things like that so I think we can't first of all when something is so important that other companies depend on it and ordinary people depend on it we normally call that a utility and that requires extra regulation we have to say yes you're a natural monopoly therefore we have to help you set a price that's fair and we have to work more that's now we as the people ask our governments to help make sure those companies behave in a fair way the thing is that they're transnational so I actually think we need innovation at sort of the WEF level at the UNESCO level to think about how do we regulate these new utilities that we're seeing there are transnational utilities and I think that's one of the things we're going to need for orchestration so we bottom up and top down both I guess right but that's regulation that's not necessarily orchestration right the orchestration of ecosystems well okay I well I guess I said the sector itself is doing some orchestrating so they aren't regulating the government is they asked to regulate right so there is orchestration coming not only from sectors also from civil society do a lot of orchestration so it's a it's a it's a bad metaphor in a way because there's not only one symphony we need diversity we have to have all these things moving around but orchestration is being done you just need to be coordinated and to cooperate with your peers that was one thing I very much liked and Lee's we need teamwork we need team cooperation and that's what the EU is doing really well that's so that we can empower even very small countries but we can also do that through other mechanisms like UNESCO like WEF thank you I think when you talk about orchestration obviously I think it's natural for everyone to think governments should take a lead and and and I agree with that especially in the beginning of you know building an ecosystem effort around new technologies and new industries what's very important to really underscore here is that government alone cannot do it on the themselves so you need that kind of partnership and collaboration that Bryson is talking about and figuring out who plays a stronger role at what levels of this industry and technology maturing because yes well government will always be the drivers seat at the very beginning but as the industry matures as the technologies become more embraced and adopted then you see more of the private sector you know you know taking a leading role so I think it's just figuring out how you do it when I did share the example of zipline I remember when they came I mean what we knew as a country was that we had so many challenges to solve for especially starting with healthcare so what zipline did was to present to us we have drones that we can use for logistical services our problem was we need to deliver medical products especially blood products to healthcare facilities across the country where the road infrastructure is probably non-existent where it takes more than three hours to deliver some of these emergency products where it will require huge investments in terms of cold infrastructure a cold room infrastructure to store some of these products vaccines blood and all of that and so we said we have a problem we seem to have a solution we don't have regulations and I think what's important to know is that not everything has to be regulated and as governments we cannot regulate what we don't know and in the technology space what we tend to see also is that sometimes the innovations the technologies are way advanced than the kind of policies and regulations that we may you know already have in place and so we had to choose the path of let's experiment with this solution knowing that we don't have any regulation or policy let's build the regulations and policies that will enable an industry a drone industry that will then look at other different use cases I know Lou talked about you know some of the healthcare solutions that I wanted to share one that we've had where we worked with Babylon to put in place an AI chatbot what was very interesting and we talk about trust we had people the patients how comfortable are they not going to see a doctor in person but really you know calling through and you know using the AI chatbot to have a diagnosis and trust that diagnosis without really seeking a second opinion that was one I think what was obvious for us was that people really needed to do it and but also not everyone so the you know middle class and above everyone is comfortable I don't want to show up at a hospital if I can call in if I can use the chatbot I'm happy not to show up at the hospital but for others they're like am I really sure I'm getting the right diagnosis I really need to see you know a health practitioner in person then we needed to put in place incentives for the health workers as well because this is now an added piece of work because if they're able to see 20 people a day and now you're introducing the AI chatbot so this means they're going to be 40 consultations maybe double the consultations that they could possibly make so what incentives do we put in place because now the workload has increased but you also want them to be motivated enough so that they're paying the same attention as an in-person consultation and then the third category of course was government when Babylon came because we had concerns ourselves as well to the extent that we're saying okay fine we're comfortable with using an AI chatbot for anyone 18 years and above what happens to kids am I comfortable a 10 year old you know using the AI chatbot and trying to seek healthcare services what measures do we put in place must they have like a guardian or someone in place so these were all the kind of concerns that we had but we knew very much that they shouldn't be concerns that limit us from testing that particular innovation and solution but rather figuring out what could be everyone's concern and how do we really rally everyone around it test it we started with a pilot we love pilots in Rwanda because that helps us to really you know establish that are our concerns real and could there be other risks that we're not looking at that are probably going to manifest when we start to implement some of these things and I think that is what really also then eventually allowed us to be able to do a scaled deployment across the country where we're able to use this AI chatbot so I think in a nutshell that's it and one last point that I want to make and I think for the innovators the business people here will agree with me I think the bigger challenge we have is policy methods but also as an ecosystem bringing in question the issue of governance and regulation is how do we streamline this because if he's in any European country and is thinking to go to Africa or anywhere else and he has to navigate 10 sets of regulations first of all if you're talking about a highly sensitive industry like healthcare and financial services where there's sensitive data and there's all these risks around it it's going to take him almost a year every time he expands into a new market so how do we allow for beyond the national ecosystem how do we create that international ecosystem that allows us to really unlock these regulations at an international level where we streamline and harmonize because if they have to go to every new market and go through a year of figuring out how to comply and how to meet the you know the requirements that are required then scale is probably never going to be possible. There's a key point here right and that is learning by sandbox right learning by experimentation space and I want to get the two entrepreneurial leaders here to comment on this but before I do let me just remind the audience that we're about to go to Q&A if you have any questions I believe our WEF team is going to manage that so feel encouraged to flag them. Both of you heard this notion of experimentation spaces right because frankly there are so many things we just don't know yet. I heard you say don't over-regulate too quickly in ways that are going to stifle my business but I also heard you being very responsible that's the essence of your business right so how does that tension that creative tension work out right I heard Lou talk about the orchestration of these ecosystems right which probably goes a bit beyond what your company does as much as you educate so what's your reaction here on especially this point of these experimentation spaces sandboxes across borders would that work for you what would be a wish list what would you want to see firstly I want to say talking about trust talking about a cross boundary but basically we need to talking about trust for healthcare they don't trust the IT company I don't think the people find out to make a diagnostic they trust the hospital the secondly they trust the regulator so for example we have a kind of software for stroke diagnostic when the people got a stroke they have a very limited time window in some hours they cannot move from a small country to the big city the small town to a big city the people will have big trouble so it must use very limited time to make a diagnostic make a searching and something so we did is we cooperate with the best hospital in Beijing number one all that the others hospital trust them not trust us we convert digitalized the clinical pathway firstly scanning the imaging and then they follow the steps of that and to say what is the next and some others hospital just follow up the you know the clinical pathway so then there's a big hospital connect few hundreds county level second tier hospital they connect together connect together they're sharing about data like a big doctor can talk with a small doctor so they build a kind of community that's one secondly those kind of software is very much a series because this is the life of people even make a wrong way the people will be died and the big trouble we must be passed a kind of check or verification from the government like FDA China called SFDA we must got a kind of a certification that is a ticket so that means after the check of that that is software is just like a digital doctor so see that the basic did the level is the same as a high-level doctor but much better than you know the normal doctor so the two parties very important I'm hearing three things maybe you could comment last but I'm hearing essentially in each one of these sandboxes experimentation and learning spaces we need some something of the nature of a kingpin for trust somebody who is inherently trusted and flow our innovation through them because they already have credit with the people right second was we need connectivity and network effects in that kind of experimentation space so you can more readily have that orchestration and that scale out and lastly we need checks and improvement and iteration right some kind of roadmap on on that kind of checking and iteration improvement in that kind of sandbox darko final comments and all that yeah I think individuals trust experts so if I'm a patient I trust the doctor more than I trust the government or the or the regulation if I am a business decision maker I may trust my my domain expert my my analyst to help me so I think the key to get AI to be trustworthy is to have the expert of the system directly feed their knowledge into the algorithm and this has not been possible in the past because we've just relied on historical data stored in some database we've completely neglected the intellect of the of the experts and I think we are very very excited about the latest developments in AI with technologies like causal AI which are able to essentially download the experts knowledge into the algorithm and then we have the best of both worlds we have the best of data we have best of human and then we have trust in the AI because we know the doctor has put their knowledge in we know that the engineer has put their knowledge in and I think that is the key to trust in artificial intelligence yeah and I suppose the aspect of localization in this it can be you know I you know I don't know the situation in Rwanda but having an American doctor weigh in on a situation in Rwanda probably would be a misallocation of his expertise or their expertise right you want local doctors that know the local health environment right yeah Joanna I'm I'm just a little concerned about the the the portrayal so I I hope you don't mind if I combat a little bit some of this I mean we were involved very much with actually have a master's degree in expert systems from 1991 but but I yes that matters but realize there that and I love the the grounding the idea of the partnership with someone you know right so someone that you know that their reputation but it's not just about you know being sure that there's a system that you've been convinced has come from some particular expert you you need to think about cybersecurity for example so even if in all good your system works perfectly in the beginning what if someone hacked into it right there are standards that we have to put in place and I think a lot of people do trust the governments it varies by country a lot so some countries have established the government's established trust and then sometimes that varies over time too but I really think we don't want to say that it's only between that we're going to you know magically be able to get from one person to one AI system remember that there's a whole always a whole ecosystem and that all needs to be managed of course I actually don't don't disagree with you on that I think that that's that makes a complete sense of the kind of a high level I was just you know speaking at a lower level where you know if it's if it's a health care AI having you know reassurance that bunch of doctors have put their knowledge into the algorithm is really really important for trust but of course I think at a high level as you build out the systems those concerns you know yeah so it's essentially expertise and governance going hand in hand right okay we're having to go to Q&A now delightfully so any any questions and if so please your name and organization sure hi my name is Amit Kakar I actually am an MD in oncology and nuclear medicine I work with novel holdings and we are active investors in AI based health care life sciences and sustainability companies I sit on the board of cure.ai which is actually working in Rwanda right now and also Accenture which is one of the AI based drug discovery company I think my question is very fundamental and that's data if you look whatever you guys need to generate today to have you know more resilient systems you need data and unfortunately every country is going into a data silo today patient data is not being shared evenly across the world today and it's not just one country it's pretty much everyone how do you make your system more robust how do I have a billion scans on an x-ray or a CT to make sure that I have a higher sensitivity and specificity when I diagnose a stroke a tuberculosis or lung cancer how do you achieve that in today's world it's a great question data is the fuel how do we get it and safeguard it Joanna do you want to take this and we'll go around the circle yeah sure I just very quickly um so everybody's worried about the data privacy as they should be it is part of the role of the government to defend its citizens and so citizens need to we need to know we know now that citizens can be manipulated right so the data is a part of your your person we need to be able to defend that however having said that there's lots of technologies now in place for even just here yesterday people were talking about doing machine learning on encrypted data right also it's important to understand that uh if there's something that's very regular so if you do actually have you know there's eight billion people with lungs you can often learn quite a lot about lungs from a relatively small number of people who have consented volunteered who have who have volunteered and consented and you don't need necessarily to have every person at every point I completely agree with that I think technologies like homomorphic encryption allow you to learn from data you haven't seen which just sounds like a bit magical but it's actually actually quite possible to do so I think emerging technologies here are probably going to come to the rescue but you know it takes time for those to scale before we can we can apply them so yeah I should share what we're doing in that sense one of the things that we did about three years ago was put in place healthcare digitization roadmap and what we were looking at and this was before we even put in place the data protection and privacy law and by there it was a very exciting exercise when we're doing the data protection and privacy law because we had to figure out what would be the right balance between not stifling innovation by being very protective versus how do you enable innovation by the same time as government we need to make sure that the privacy and the protection of this person of data is really you know taken seriously and what we did I do agree with you I think what you tend to see is across different healthcare providers everyone has their own data it's siloed as you said but also it's very unstructured then the accuracy of that data is also in question and so two things were very key in our health digitization roadmap one with how do we enable for the random population a single electronic medical record so what that means it requires that we harmonize across all healthcare practitioners so I have one record so regardless of how many you know health practitioners I'll go and see I still have one medical record that I can enable for anyone to see but also ensure that I have all this information stored in one place the second thing was to put in place a health exchange information platform which would allow then for all these siloed you know hospital management systems and data states that are in place to ensure that then you know you're able to exchange that data across the board and then finally I think without data protection and privacy law as a citizen you have agents you're the custodian of your of your of your data you then decide who looks at your data who sees it and who you get to share that with and I think it's going to be some you know at least three years in work to sort of get us to the point where we'll have one medical record per citizen per patient and then enabling that consent that the citizen has over their data and I think after within the next two to three years we should start to see the ability to then now how do we anonymize this data and open it up to to players such as yourself the private sector so that they can innovate around that data because clearly and I think we all agree without data there is no innovation and there is no advancement on the human condition we have to step away from data being a dirty word and we also have to protect that data privacy assured agency assured before it changes hands and sometimes we might arrive at models where the ones that we heard yesterday where you might not actually need to exchange data because you can send algorithms there even encrypted data can now be read very important topic the global data economy incidentally measured at forecasted about six trillion dollars annually so it's a huge opportunity for human growth if we ethics assure it any other questions here let's see who else yes sir hi my name is Sean I'm a founder of college intercom which is a platform that we help students to use AI to create their own study plan and we're just confronting the same like trust problem problem as Mr Liu just mentioned a quick question for Mr Liu do you think the trust problem is just a time problem or it cannot be solved through time I mean like when when internet was just worn a lot of people don't trust internet like at the very beginning but like we're have been already very used to everything through internet what do you think about AI thank you I think that is just a timing problem even we we face many kind of a challenge but AI can really help the social development it's true if we look at they can help us to very much improve efficiency of our works and sometimes like like I mentioned about doctor you know the doctor is very much individuals so if we we use AI it's very low level requirement it's not like best doctor become more excellent we like a low level not low level doctor it's not the good you know good experience doctor 10% better than before that is a great contribution to to the cause of healthcare and the people got more happy for that so AI have so many applications even today personal assistant already help us solve that problem this timing will solve a lot of problem like a data privacy I'm so happy to say it will come each of country developing country or developed country every country already starting to make their data privacy you know governors and something everything they already start not only today it's already a few years ago so that means it's just like you know when the technology move ahead and then the regulation will come in maybe a little bit later but we need to give enough space like technology like us like a technology you know the innovator gave us a little bit of space like us to find that something that never had before so I think every people should be happy for this kind of very powerful digital world computing power connectivity and also the data it's it's a time data is very low cost so that is a good time for us I might want to augment that maybe Joanna you want to pick up on this but the element of time of course there's optimism baked into your your answer here Lou of course as an innovator that we will figure this out as time goes on but I think we can also agree that if we let this slide for too long we accumulate more what we've started to call tech debt which means too much data out there too much violation people stop trusting then government comes in with an overly heavy hand so clearly as time passes we have to be thoughtful about how we design these systems and that is maybe how AI is different from other revolutions beforehand well I again want to say that sort of defend how good a government is is something that the people of the nation and the government itself helps determine so it isn't necessarily that eventually the government comes in with too having a hand after having been too libertarian for too long sometimes they've already lost control and they can't come in and they don't know what to do and some other nation winds up doing the regulation or some other group of nations so I think we've seen these kind of dynamics but we can also have these ideal situations like what we're hearing with Rwanda where they're working really they're working with the international it's not easy it's not easy dealing with some of these big companies come in that some of them have really exploited and taken advantage of developing nations locking them into 30-year contracts that were were disadvantageous so terrible terrible things have happened but some countries are are are doing a good job of of threatening that needle of working with large tech working with you know supporting their infrastructure and really you know people talk about knowing I know even going between different states of the U.S. that some states have good senators that are responsive and some states have senators that the that people can't talk to right so you realize that one of the things we need to do what we need in a good world or to get things working is to be in control enough control of our government that we can trust it and then recognize that when you have problems sometimes you have to move and work in these other institutions but it's in everybody's interest to work collaboratively thank you Joanna and thank you everybody I now have the thankless task of trying to summarize this very exciting conversation which is so multifaceted I think there is just a maybe a couple of very high-level takeaways and then we'll summarize this and writing for you all afterwards clearly government is here to stay and it should be here to stay as the guarantor of doing right by people that means economic prosperity it means social justice it means human dignity right so government is here government needs to be brought into development and we also have to recognize government alone can't do this it lacks the expertise government knows it lacks the expertise therefore tighter collaboration within these experimentation spaces and I think that is sort of the key takeaway here that readiness is essentially about experimenting with what that actually means in a thoughtfully designed way right so as part of that we need to have the trusted entities we need to have the network effects we need to have the cross border and we need to have the trust assurance none of that can happen without government but government itself is not going to do the trick just like we've seen corporates and startups themselves can do the trick it's about that tight integration so these parts that we normally perceive as readiness which means infrastructure and education and training and skilling and data pools etc they're all critical ingredients but they're nothing without a thoughtful design and we don't have all the answers now we need to experiment ourselves into those hand in hand to do right by our societies so with that let me please thank our esteemed panels here that was a fantastic conversation thank you very much and of course our audience thank you and we will see you again here soon