 I'm Jennifer Shanker, editor-in-chief of the Innovator Global Publication about digital transformation and sustainability. A warm welcome to all of you joining the panel on fast-tracking progress through data and AI. I encourage participants to post their questions to the chat. If time allows, I will pose your questions to the panelists. Our focus today is on how to best apply data and AI to the UN global goals. The potential benefits are well-known, but there are a number of challenges to scaling the technologies. For starters, we know the implementation requires a significant amount of data, which often sits in different hands. And complicating matters even further, the data may be stored according to different protocols. Data analysis requires significant computational infrastructure, which can be cost-prohibitive, as well as communication networks to access the infrastructure. The AI models used can be proprietary or expensive as can the tools needed to build new ones. The countries that need to use AI for the SDGs or the organizations working on them may not have the skilled engineers to combine models, data, and infrastructure in an optimal way to produce an output. And even if they can, once an output or an insight is generated, it's only as good as the management or domain-specific plans the organization or government have to address a specific problem. So to summarize, there are five barriers to scale that need to be addressed when we apply data and AI, and these barriers are particularly acute in emerging economies. Data compatibility, meaning the ability to make enough siloed data accessible for model training, access to computing resources for the purpose of training AI models, access to top-tier AI models or to the tools and libraries needed to build new ones, technical expertise and skills to integrate data, computing resources, and tools and models to produce insight. And finally, the domain expertise and management capabilities to turn AI-generated insight into climate action. The goal of this panel is to look at some concrete examples of how data and AI are successfully creating impact and share some of the lessons learned so that more organizations and more countries can overcome some of those barriers. I'm very pleased to introduce our panelist, Charlotte Petri Gornitska, Deputy Executive Director at UNICEF. Welcome, Charlotte. Paula Ingebiray, Honorable Minister of Information Communications and Technology at Rwanda's Ministry of Information Communications Technology and Innovation. Ocean Data Technology Champion, Kimberly Lynn Matheson, General Manager for Microsoft Norway, who will assume her new role as CEO of the Forum Center for the Fourth Industrial Revolution for the Ocean in the coming months. I'll begin by asking all of the panelists from their own experience and knowledge to share one or two examples of data and AI being leveraged effectively for the UN Global Goals. And then to identify which of the examples you mentioned, you think will have the highest potential to scale globally. So let me start with you, Charlotte. Tell us what you think. Thank you. I will try to be as concrete as possible. Thank you, Jennifer. But first, I agree with the five elements you mentioned as a framework, but UNICEF and I would be inclined to add a six on responsibility. Three years ago, in partnership with NYU's GOV Lab, UNICEF launched the Responsibility Data for Children Initiative. And that partnership has a focus on field practitioners and it's all about guidance, tools and leadership to support the responsible handling of data for and about children. And I'll put in the chat box later where you can find out more about this. It's also important secondly to talk about the access to digital solutions, including those that leverage AI. And we know they are severely restricted for much of the world, especially for those who stand to benefit the most from these solutions. Of course, this is a driver for UNICEF. And we talk about the digital divide when we talk about AI, it's even more so. So we think that's very important. A colleague of mine in UNICEF says, UNICEF cannot rest until we know that the last adolescent girl with a disability in northeast Nigeria has access to the services she needs. And what I appreciate most about this statement is it doesn't only underline what the Sustainable Development Goals talks about a lot, meaning leaving no one behind. But it also accentuates the need for appropriate disaggregated data. And we need AI and machine learning to support this. And in terms of an example, Jennifer, you asked for examples. Last month, UNICEF launched the Children's Climate Risk Index. And this index uses data to generate new global evidence on how many children are currently, and we talk about that, but this is data. But how they are currently exposed to climate and environmental hazards, shocks and stress. And it shows that one billion children, nearly half of the world's children live in a country at extremely high risk of the impact of the climate crisis. We're talking about 33 countries and they just contribute 9% to the green gas emissions. So there's a very important data here. But let me, I think that the index is important as such. It is bringing together geographical data by analyzing exposure to climate and what I talked about hazards, shocks and stresses, but also in combination with child vulnerability. So this index helps to understand and measure the likelihood of climate and environmental shocks or stresses leading to erosion of the development, mainly for children and vulnerable household groups. And I'm not excited about the kind of figures that we see in this index, but I am excited. Because the index uses data science analytical techniques that are kind of new in synergy with existing statistical data and analysis. So bringing together established and new ways of working can actually realize the benefits of both. And lastly, what we do see, we talk about scale up, and of course we need to scale up solutions. But when we talk about data, we actually need to scale down or scale out because we need to know exactly what we need to do in northeast Nigeria for the adolescent girl. And for this purpose, we need to be much more knowledgeable about that situation and we need disaggregated data. So, thank you. Thank you, Charlotte. It's clear that this index can have an impact because it actually measures the impact on the children and hopefully that will be a catalyst for action. So I'll turn now to you, Minister, and ask you to give us a few examples of how you see AI and data impacting the global goals. Thank you, Jennifer. I'm happy to be joining this panel with both Charlotte and Kimberly. And I'll be very quick because I am being mindful of time. One of the examples, one of them being how Rwanda as a country has used data and AI pre COVID to respond to some of the, you know, challenges that we were seeking to find solutions to as a country. One example I'm sharing is of a local company called carries that was using drones to to in the fight against malaria and what they were doing, we were using drones to capture area images, and some of the marshlands across the country, and would use that to develop an AI model that could quickly detect the mosquito breeding hotspots. And this allowed us as a government to quickly up to the information and intervene with more targeted widespread spraying using these drones. So not only were we using area images to figure out these mosquito breeding hotspots were also using the same drones to then spray these marshlands and prevent the spread of malaria and I think this is a clear model to your point that can be scalable, especially to most parts of the developing world because we still have malaria that is prevalent in most parts of Africa. The second example I'd like to share is a partnership that we had with the GSMA where and this was during the COVID times we're looking at the changes to public transport capacity due to some of the preventive measures the social distancing requirements that we had put in place. And as you can imagine, the, you know, the transport companies or transport operators their revenue is reduced significantly because they, their current capacity was at least at 50% and on those good days will be around 75%. So that led in a mismatch in both the demand and supply of public transport. And so what we did was to look at some of the ticketing platforms that we had in place and I think this is really a great example where we realize that what has happened is that we've invested in tools that allow for bus ticketing and so you're able to use this data, you're able to use call data records as well for citizens and this led to us being able to figure out bus optimization, route optimization, which routes are mostly congested now that the supply for public transport has significantly reduced given the COVID 19 preventive measures that have been put in place. And more excitingly what we're able to churn out is with this new with embarked on a new agenda of, you know, advocating for immobility and putting in place incentives to have the different immobility providers whether it's for motorcycles or cars. And so this same data has enabled us to sort of like map out the different charging stations where they should be placed and in turn attract the private sector to to place some of those. And so these are two concrete examples on how, as a country as a government and together with the private sector we've been able to leverage the power of data and use that to inform decisions but also to build models that can then, you know, sort of feed back into some of the interventions that we are undertaking, opposed there. Thank you so much minister and what I love about your examples is, you know, first it underscores the importance and potential effectiveness of public private partnerships, but also that you're using AI and data not only to identify problems but to actually take action. So, you know, great examples, thank you so much. And now let me turn to you Kimberly. Yeah, tell us a little bit about how data and AI can help us save the oceans. Great. Thank you so much for having me here I love the examples given so far. Three great ones and in fact I love that you start and ask us to get pretty specific on examples so I'll get right to that but let me just make the comment to say what what we are in fact, because to the five and six challenges you just mentioned we agree that getting data for our oceans are getting data for just about anything gathered together and the aggregate that we really really need is extraordinarily difficult right now. And this is the motivation, particularly around oceans that World Economic Forum with the Center for the Fourth Industrial Revolution the awkward group and Microsoft came together in order to form the world's. Ultimately the ambition would be to have the most important data collaboration platform highways uniting all of the data we need in order to provide the insights to the world where they need it so that each and every nation, each and every organization doesn't have to go out and try to build that themselves so that is exactly the nature of what we're trying to do in scaling and on a basis which has never been done before. Now to that end, we have imported, we've built a platform we've imported an awful lot of important data already and I'll give you a couple examples of how that's already working in pretty exciting ways. In the first instance, we, we need to promote a whole lot more shipping on our oceans, and a whole lot more aqua culture to take two examples so food from the oceans in order to be able to paint a good word going forward but we have to do it without destroying the oceans. So, what we've done is look at shipping first if I take those two examples shipping first and say, we've combined data from an automated identification system it's it's fairly kind of partly open data today it's a system that tracks 200,000 ships worldwide moving around. We imported that data compiled it together and fed it into a recognized model for for what the emissions were likely to be, you know, ships across the world right now are largely not equipped with the sensors where we can easily pick up how much emissions are going out. Well, we can wait a long time for that to get fixed, or we can use AI and data that we already have available in order to model with a pretty good level of precision 200,000 ships that are out there already today and we've done exactly that. We've put that data out there, and as we now start to bring that data to life we get a whole lot of interested people and helping us to make that data even better, so that that data actually you know interested parties parties that are running those ships, and those transitions for example really wanting to make sure that that data is accurate that we do have transparency around it right this is to help compliance but this is to also help ultimately catalyze a transition to cleaner fuels and be able to highlight where companies are taking responsibility and where that's going well and give credit where credits do and usher in change faster. So it's a fantastic example. It's a great example and if I'm, if I'm correct. I think that the data shows that the emissions are actually going up. And, and so if action isn't taken, things are going to get a lot worse. So, tracking this is super important. It is you're right it's super important do you want me to share one more example or do you want to do you need to move us along for time. Thank you for your other example. Aquaculture is maybe, you know, another interesting area to focus on where we see, for example, issues of lice on fish and issues of algae blooms. We've seen terrible losses in the industry because of these issues in the aquaculture industry so where we're raising salmon for instance in the oceans. We're going to get proper learning out of what could prevent those conditions from happening what could prevent that devastation of those crops of fish right which is bad for the environment bad for the economy it's bad for everybody. We're finding out through collaboration that the industry themselves is is very very interested in forward leaning that they're sharing data now in a way partnering in the early stages now but partnering with our ocean platform to get data. We're going to get to the platform and get these different fields which are raising the fish collected together because each one has trouble demonstrating all of the causes and then acting on them but when we aggregate that data across very many different places across very many ocean ecosystems and many ecosystems across the planet suddenly, we can learn a lot more, and we can do things that are good for everybody to remedy those conditions and spot them and get proactive about knowing much more to to head them off if you will, or to minimize the damage and then maximize the outputs from from our endeavors. So another great example the world needs. Yes, I love the example of aggregating data because we will need cooperation between various entities to to really fully leverage the power of AI and data for sure. So, you know I'm conscious of the time, let's move into the second question about how can governments and organizations overcome the challenges related to data access and resources and infrastructure for training the big data, big machine learning models that we encounter across different climate use cases. And I'll start with you Charlotte. Thank you so much, I think what we can agree, all of us is that AI and data talent and breakthroughs today have very little to do with the most vulnerable and the STG so we need to turn that around and partnering around. First of all, one of the things that we've seen work in in partnerships before is when agencies organizations like UNICEF and others who actually know a lot about the problem. Share the problem with those who can really work with the solutions and that has to happen. The private sector thrives from problem solving. So we need, we need to get the people who are now perhaps in Silicon Valley by themselves and the UNICEF first. I'm sorry if I use UNICEF as an example but just to be very concrete, we are there in our own ecosystems, we need to get them together. The problem in the driver's seat we really need to make sure that that we invest critical mass also closer to the use cases, so that we can get this aggregated reality as well as data, and then we need to share. There's a lot of data we don't share so we need open source and transparency. So that's a few examples from my side. Thank you. Thank you so much Charlotte. Let me turn to you now Minister because I know Rwanda is tackling some important issues through legislation. Can you tell us a little bit about that. Thanks Jennifer and maybe I'll start off on something that Charlotte pointed out which is really sharing data. And that's one of the challenges when it comes to for many of you know the organizations that are leveraging big data. To even build their AI models where they lack access to data. So there's one aspect of we don't have enough data or we don't even have the data that we're looking for. But there's also the other aspect of the quality of data because if the quality of data is lacking, obviously you can imagine what the output of the insights will look like. So I think that's the very first challenge that is related to data access. And then the other one is having in place the right governance instruments to make sure that the data that is used is used in a responsible trusted and secure manner. And so making sure that you have, you know, some of these instruments is also, you know, a key challenge that I see largely. And today we're working with our Center for Fourth Industrial Revolution which is an affiliate center of the World Economic Forum to address some of the governance gaps in this area to unlock the potential of data. And we are also looking under this Center for AI Policy and Innovation, what we want to create is a national AI research cloud that can host high value open data, open government data sets for research purposes. And this is substantiated with other government interventions related to availing data sets that could be used to develop AI use in public governance as well as supporting the research community. And because of that we also put in place about two years ago we put in place an open data policy where we were encouraging most government institutions to really open up some of their data sets, whether it's for innovators but even as we think about building this AI industry knowing that it's banking heavily on the availability of quality, reliable and available data. And it was important that as a government we even take a lead we walk the talk and really also opening up some of the data sets because we happen to also create relevant solutions. And as I finally, my last point on this is we're putting in place an AI legislation, which really comes into address challenges around access to data around how do we get the right skills that are needed the expertise that is needed to be able to analyze and synthesize and be able to provide tangible insights but also really figuring out how do we attract some of the companies that are already building AI capabilities and solutions to come and consider Rwanda as a proof of concept place where they can test and try some of their AI models and solutions and if proven successful we can then think about scaling them to other parts of the world that are grappling with the same challenges we're trying to address with these solutions. And I think these pieces put together what will really allow us to have a thriving AI industry. Thank you so much for outlining that for us and I think it's clear that you know you have our laying the groundwork for Rwanda to become a test bed for all kinds of different AI and data applications. We're quickly running out of time but Kimberly I want to give you a chance to weigh in. Can you just quickly tell us like, you know, the one or two things that you feel are needed to overcome the barriers. Yeah, in the first instance. When we listen to the minister from Rwanda that's done such an impressive job of really being an advocate for sharing data and very many great things. It also just reminds me that we don't want every nation out there building this themselves. So one thing that I think is extremely powerful. For example, the Norwegian government has spearheaded a high level panel for ocean sustainability. They're in a silver government 14 nations are a part of that already. And that important working group has decided to give the mandate to this sent to the center that I will become the CEO of the Center for the Fourth Industrial Revolution Ocean in order precisely not to have every country building an asset a collaboration and apparatus which uses AI makes these insights readily available. Again and again and again out there. So that's a fantastic scaling mechanism for all of us to get to enormous important value earlier. And I in my last comment I want to make is, we're going to work really hard in our center as well to open up the industry data, much more broadly, and much quicker than it's been opened up so far and that's the uniqueness of what we want to bring to this as well to literally create the determination and the compelling cases for industry to mix their data together with government data with research data and be able to have that interplay happen in a way sooner than ever I could talk more in the breakout time afterwards on the barriers to doing that we're learning a lot so far. But but that's that's an incredibly important activity and you know we're on the case it's going to be a central theme for us. Okay, fantastic we are. We are almost out of time I want to thank our three panelists for giving us some really terrific concrete examples and and of things that you've seen in the market but also about how to overcome the barriers and these these will serve as a great jump off point for further discussion.