 Thank you all very much for coming today. My name is Alan Dengore. I'm the director of Climate and Health at the Welcome Trust in London. I'm really pleased to be here today to kick off this discussion about data and digital innovation and really pleased to be doing this in collaboration with our colleagues in the Rockefeller Foundation. So we're here to discuss how data and digital innovation can really transform this field of climate and health. And before we get into that, I just want to say just a few comments about what we mean by the field of climate and health, because it's a new field. And it falls at the sort of borderlands, really, between existing fields of public health and climate science. And that makes it a particularly challenging field to work in, because doctors and public health people want to talk to public health people and other doctors. And climate scientists are very happy talking to climate scientists. So the difficulty of getting those two communities together, those two disciplines, those various skill sets together, is really quite large. And the difficulty is particularly acute in the area of data, where we have completely different understanding of data. So the climate scientists understand what's going to happen over the next 50 years at a stratospheric level. And their health people understand what happens to an individual over the course of their lifetime. So we have different types of data collected in very different ways. But before we get into the data and what we're trying to do, welcome, and others, let me just talk about those interconnections between climate change and health. So we typically think of three different pathways. The direct pathway, it's getting hotter. There are more extreme events. And this obviously has immediate impacts on health. The pathway is mediated by nature, by the environment. So as the environment changes, so things happen in nature, which in the environment, which, for example, the distribution of vectors like mosquitoes, disease vectors like mosquitoes, those mosquitoes can live in different parts of the world and are. And we know the evidence is very clear that the distribution of mosquitoes is increasing enormously, which will lead to a change in the distribution of infectious diseases. The ability to grow foods as the environment changes, of course, is another example of how the environment interacts with human health. And of course, we know very little about some of these interactions. And mental health is also part of that discussion. And then the socially mediated pathways. So that's as the environment changes. So society has significant challenges. And that can lead to, for example, migration, can lead to social breakdown, can lead to conflict, and at the extreme event, and at the extreme ends. And that's obviously going to have significant impacts on health as well. So those are the pathways. And those are the sorts of pathways we need to interrogate with data and with innovations in digital science. And what are we doing in welcome? Well, in welcome, we are seeking to put health at the heart of climate action, to really use health as an argument, as a fundamental argument to driving action on climate change. So understanding the impacts of climate change on health, understanding the benefits to health of rapid climate change mitigation, understanding the adaptation actions that countries or cities can take to protect the health of their population. But underpinning all of that is a much better understanding of the analysis, the approach, the methods, the tools, the data sets that we really need in order to conduct high quality analysis in this area. And at welcome, we split that into three things. The first is the data sets. How do we bring together the climate data sets and the health data sets in a way that enables us to do the sorts of analysis that's required? The second area are the sort of the tools, the methods, and the metrics. How do you, what is the best metric for an impact of climate change on health? How do you attribute, actually attribute the impact of climate change itself on health as opposed to poverty or as opposed to other things? And then finally, the third area, of course, is we need to think about equity. And we really must think about it is not appropriate. It's not at all that very few people, largely sitting in universities in the global north, are able to do this analysis. So building, strengthening, the cohort of data scientists and data professionals who understand and who work with and who innovate in this area is going to be absolutely critical and something we welcome are totally committed to. So with those introductory remarks, you don't really want to hear from me. You want to hear from our fantastic panel. And I'm absolutely delighted to hand over to Raj Kumar, the president and editor-in-chief of DevEx to take us through this session. Thank you very much. Thank you so much, Alan. And thank you for not scaring us to death. So Alan and I met in Charmel Shake a couple of months ago. And I walked out of that conversation a lot more shaken than all of you look from his introductory remarks. And I think for good reason, but you presented it academically, dispassionately, and objectively. But there are some scary facts here when we think about the nexus between climate and health. I also think of this session as the quintessential Davos session, partly for the reason you described. We're talking about two leading themes in the world, climate and health, and adding in a third, which is digital, and layering them together. It's a kind of thing that only happens here in these ice villages. So I'm excited to have this conversation. And the folks at Welcome in Rockfeller put together a great group for us. So let me welcome to the stage Carl Brinton, who is the managing director of strategy for the Rockfeller Foundation. Please have a seat. Joanna Shields, the CEO of Benevolent AI. Please. And Tallulah Oni, who is the public health physician scientist at the University of Cambridge. Tallulah, please have a seat. And we're going to have a conversation about these three themes and how they connect and where we go from here. And hopefully include a number of you in the room. So if you would like to jump in, take some notes, think about what you want to say, provoke us if you would, tell us what we're missing. Maybe we just start at the high level here. So it seems like the time frame is a key point. Alan brought it up that we often hear about climate targets as 2050 or 2100. And we just wrote a story in DevEx in which Alan has heavily quoted that does do a bit of the scaring I just talked about, where it says, look, there's actually quite a bit of this could be happening right now and within the next decade. We may see many of the effects here, whether it's extreme heat or the movement of disease vectors like mosquitoes. I guess we'll be just starting at this high level. Are we at this point where if you're in government, if you're at the WHO, if you're in a not-for-profit working on health, you need this data right now to be tracking? Is this an immediate issue? Or is this sort of like a let's look at into the future as we often do at WEF? And imagine what it's going to be like a decade or two decades from now. Maybe we just get a quick take from all of you, and then we're going to dive into many more details on this theme. Carl, go ahead. So in short, it's now. And a little bit more on that. If you don't just ask people at Davos, if you ask people in Nigeria, if you ask people in Pakistan, if you ask people in Queens, so I'm in New York City. And a year and a half ago, we had torrential rain. And people died. People died in New York City, in the global north, a number of people. This is real for every single person on the planet. It doesn't matter where you are, and it's real right now. Joanna, what's your feeling? Is this on OK? Yeah, I think the first time I got exposed to this, I was a government minister in the UK and bringing digital into various government departments at the time. You were a UK minister of internet safety and security. This is another topic. But we had a flood event in the UK. And we had once in a generation type flood event in 2015. And it was like we called in people from the technology community and said, we've got to be able to warn farmers to move their livestock. And how can we do that with ordinance data and all kinds of other things? So it was like a big data science project. And it became clear to me that the work that we do in technology development and the use of AI and machine learning for data for all kinds of solving big problems, that it was going to become a real issue relating to climate. So ever since then, I've been focused on these things. And I was the co-chair of the Global Partnership on AI for the last two years. And we did a big work stream, brought experts together, and did a big work stream on responsible AI for climate change. The health angle, we didn't bring those two together. And I think that's a really good next project for G-Pay. Sure. Well, what I like about your example of the livestock is it shows if you had the data in advance, maybe using predictive analytics. If you knew something's coming, it's going to affect this area. And if you had the ability to reach and pinpoint the farmers affected, you could actually make a change right away, a change in advance of a disaster. Maybe we'll just go to Totaloo quickly if we could. Your take on, is this an immediate issue? Is this a bit off in the future? So yeah, so you've kind of spoken about something like flooding, which is an immediate and very acute. So I want to kind of flip to the other side and talk about the more slow-motion issues, because part of the issue, I mean, the answer is yes. Of course, it's immediate. But our risk perception for the more slow-motion health emergencies that emerge as a result of climate and are accelerated by climate mean that we don't have the level of action. So if you take, for example, something like air pollution, you don't get the same level of action you get for flooding or that you should get for flooding for a high air pollution event. But if you take it from the side of the health care sector, and you look at, you would find actually spikes in hospital episodes in terms of admissions, in terms of outpatient exacerbations of existing conditions during those times. So it's urgent, but you often don't, it is not perceived as urgent, because there's a little bit of separation in time and sometimes space from the very acute health implications. And your expertise in particular is looking at urban health and studying how, especially rapidly growing mega cities, particularly in the continent of Africa, how can they design policies. So let's take the air pollution example. If we did have the data, if we did have the digital tools, what might a city, a mayor, do differently knowing that air pollution is going to get worse in a particular area? So I mean, in the cities I work, I work predominantly in, as I mentioned, large cities, primarily in Africa. So I work in, it's like Lagos, and you don't shy away from what you mentioned. Somewhere between 20 and 30 million people and rapidly growing, Accra, et cetera. One of the most striking things is how little is measured to start off with. So you say, what does the policy act to do with air pollution? Well, firstly, it needs to be unearthed. Something like air pollution is also normalized. So it's kind of, oh, you know, they're air today, what are you gonna do? So it's normalized. So the first thing is kind of making it obvious that it's not normal, right? We shouldn't accept accepting it. And a really critical part of that is measurement, but not just measurement. Measurement and having data in the hands of people. So one of the things that we've done, for example, is I mean, this same cities that I work in, the median age is 18, 19, right? So three quarters of people under the age of 30, 35, right? And who are really passionate, actually, about environmental justice, about health justice, about climate justice. So one of the things we did, for example, was work with young people 18 to 35 to say, can we be part of unearthing? Why it's important to measure? Because that's the first thing. You have lots of emergencies. Why should they prioritize measurement of something that no one's obviously dying from right now? Quote unquote. And what we did was worked with the young people, so we did have this running intervention. Okay, pun alert, we ran an experiment. So they designed running routes and they ran through the city with wearable air quality sensors. Firstly, highlighting the difference in air quality within the same city, so they designed the running routes to capture what they thought would be different parts, different levels of air pollution. The second thing was to highlight how little air pollution monitoring was happening in the city. So you could argue, well, what is the effectiveness or what is the accuracy of the wearables? Like, well, let's calibrate it then. Oh, that's right, there isn't any, so let's do that. And then the third thing was understanding how that could drive action. So with that majority demographic, I don't like calling them young because that sounds like a niche group, they're just everyone. So with that majority demographic, they then designed and say, okay, these are what we see as the priorities, bringing them together then with the policy actors to say, okay, what are they doing and how can they be part of the decision-making in shaping the decisions around the cities? But critically, so I mentioned, measuring having data in the hands of people and then evaluating. I say health doesn't trickle down from good intentions, right? You do the thing, you identify it needs to, something needs to change, has it changed? And they play a really critical role of a majority demographic in saying, we wanna work with you to be part of understanding what this climate and health risk is, but we will also keep you on your toes and we will also celebrate when it works and come back to you when it doesn't work and really being part of that solution. And that's the power of data that is generated in close collaboration with both the beneficiaries and the people effecting change and the people that would be getting buy-in in terms of effecting more community level change, whether that's around waste management practices, et cetera. So really underlining the public in public health, really using data to engage the public in a way to change the environment for their health. I guess, Joanna, you're unique because you have this perspective as an expert in AI and the technology side as someone who leads business, but also who knows the government side of this. So just taking this example a bit further, imagine it's in Lagos and temperatures are rising, extreme heat events are rising. What can we do today if we had the right data, if we had the right digital tools, what could we do today that might change the trajectory for health during extreme heat events in a city like that? I think the most important thing is to look at how do we scale solutions? Like every country, every city is coming up with their own plan. And that's why, I'm a big believer, I know everyone talks about multi-stakeholder organizations and some of them are really effective and they're all well intended, but some of them are less effective than others. And part of it is that we have to organize for action. So one of the things that I've found that works really well is coming up with a model national response to a problem and making sure that there's contributions from everybody in the data community and the technology community to bring their solutions to bear on that particular problem in a model national response that can be scaled out to everyone. And that's the only way, because otherwise everyone comes and recreates something themselves. And there's just no, I mean, whilst I'm creating one thing, someone can be doing something else, you see, we need to share the load. So an organization I found called the We Protect Global Alliance does this to combat using AI machine learning and other technologies to combat online child sexual abuse and exploitation. But that model really works. And so when you have a multi-stakeholder problem like this, if you bring people together and you can invite technologists and experts to contribute projects and code and ideas and then get that out to everybody, that's the most important thing. All the talking shops we have are great and they make us feel a bit better. But, you know, and everyone's got great intentions, but we really need to translate it into action. So zeroing in on a specific problem is the key here. Yeah, I think so. And Carl, I guess from a philanthropy standpoint, you're a Rockefeller, obviously you have been welcome for putting this event together. There's a lot of ways you might approach an issue like climate and health. You might approach it, let's say, by thinking about how the policy environment should shift, right? But it seems like you're particularly focused on this idea of data and digital. And I guess the question is why? Why is that the barrier or the opportunity if we could unlock it? What might it do for us that would be different? Anna, you're right. We are very focused on data. Part of that comes from our long history. So we've been around about 110 years back before data was data, data was a thing. But at the time, we talked about science-led philanthropy. And that then naturally led us to data-led philanthropy. And so we see data as a strategic function that plugs into and supports all of our programs across the different topics that we do on clean energy, food, health, finance, all of those. You know, I loved how you started this discussion, Raj, where you're talking about, you know, we've got climate, we've got health, we've got data, it's like a Davos bingo or something, you know? And we hear a lot about the poly crisis, right? And maybe it's a funny word, but it's true, right? The reason why people are using it is because it's true. We see it in climate, we see it in health, we see it in war, we see it in food. What we haven't heard yet about is the poly solution. And again, apologies for the cute word. You can call it whatever you want. You can call it the Davos sandwich. You know, you need all the layers and then it's good. But the fact is we need integrated solutions. We need to be applying data solutions to climate in the context of health. We then need to be bringing food into it. We need to finance the whole thing, right? And then we need to collaborate. And this is a point that I love whenever I interact with our data team or any of our partners in data.org or data kind or a number of different places that we've been working in the tech field, tech for good. One of the things that tech has done particularly well is collaboration. Think of what field has a GitHub equivalent where you can go and you can see someone's code. You can see, you can verify the code. You can say this person can do this job well, right? It's competency-based, that it's open source, that it's then a collaboration engine. We need more of these types of things in other fields. We talk about collaboration and philanthropy and we're really trying to do it. But at the end of the day, we've got a long way to go. And so one of the things that we've started doing recently, Rockefeller, is implementing some sort of agile structures to some of our teams. So using what tech has done to collaborate more effectively to get things done. So that's really where we come from and why we are seeing data as something that is not just a fundados buzzword, but is absolutely core to everything that we do. Maybe we can come back later in the discussion to just at what level do we want to engage on this issue? Because it does seem like where you and welcome likely engage is at this very global level. And what Joanna is saying is, hey, we got to get really specific. We got to narrow down to this particular location and problem. And let's see if we can play with that tension during the discussion today. But totally, maybe going back to your example of the wearables. So you said, in a lot of cases, we just don't have the data. And obviously, we're not going to get to the point where we can use advanced tools if we don't have basic data. It's all based on how good the data is. So what's the state of play right now with the data available at this climate health nexus? Where do we stand? Start with you. I just want to hear your sense of where we are. I maybe want to take a step back and deconstruct data a little bit, right? Because that play between what is sense and what is nonsense. And what is, if you think about a matrix of sense, nonsense, scientific, and lived, or practical, there's some things that, in the scientific community, because it was rigorous and because it was done using validated methods, et cetera, it's scientifically sense. But it's so disconnected from the reality that, in terms of the ability to inform action, it is practical nonsense, right? So I want to make that distinction, because when then we talk about how we collect data and what exists, there's quite a lot of innovative tools that we can use now in terms of understanding the environment in terms of remote sensing, satellite imagery. And so at least get a sense of variations in that could be green space, that could be heat, it could be air pollution, even remote sensing, it could be land use. So that gives us some clue as to the variations within and between cities. When it then comes to making sense of that, and what it actually means, my experience is that often the tools that we have available to capture this information are not necessarily adequately tailored to the environments where they're most needed, right? So you can have a lot of data on the food environment. Let's kind of take another that has been collected using validated tools. But then when you try to make sense of it, in the context of a city that is changing very rapidly, where the food environment is predominantly actually most people get their food from the informal sector, which is actually, again, the majority sector, how does the data that you generated using very scientific methods, how do you translate that into action when it's not contextually relevant? So I think it's important that we are, and this is why participation and co-design is so critical. From the perspective of air pollution, for example, there is very little measurement. On the city like Lagos, there was essentially the US Embassy. In the city that size, there was a World Bank project a couple of years ago that partnered with the government and installed six equity sensors, which kind of increased the number of sensors by a factor of five or two, six. And so what is powerful is that through the wearable, so again, coming back to this particular is just an anecdote, but with this particular project, it caught so much attention because the majority demographic made a lot of noise about it that it actually led to a collaboration with the government and a grant that then resulted in 30 sensors being deployed. So we can talk about where things are now, but actually one of the really overwhelming experiences I find is how much appetite there is at the city level anyway. A lot of the governments are working for change, and what the key resource challenge is often is hardware and is capacity strengthening to monitor over time. So you have environmental sectors that have that capacity to monitor, but because they've not had sensors in place over time, that skill set to do that over time is. So when you actually listen, we wrote this grant together because we said, OK, listen, what do you need? And they said, listen, you academics, you come in, you swan in, you do a bit of research, we go. You don't do what we actually need. And so, OK, what do you actually need? So we need sensors on the ground so we can actually measure. We need to stay, and we need to train our people to monitor that over time. So then we were able to say yes, but we would only collaborate with you if you bring in the young people as well, and that data needs to be available real time because that's not government data. That's public data. It's your air that you breathe. So it can be really powerful understanding, unearthing the gaps in the data. And in a very short space of time, actually being able to do something about it. So it seems to me like the tech exists, right? Remote sensing capabilities, satellite imagery, where it's advanced tremendously. And it's getting cheaper and cheaper. It's more available. But the key point you're underlining is we might have the technology, and we might have institutions that could potentially deploy that technology, but if it's not connected to what Joanna earlier talks about, that solving a specific problem, like getting very narrow, then we end up with plenty of data, but it's not actually useful to us. Yeah, exactly. And I would highlight two other aspects of data that would make things more useful. One is the integrated data, integrating the health data with the environment data, with the climate risk data. Because again, really powerful, just being able to see the, we're not talking about long-term impacts of climate and health in the next 20 years. Just over a month or two period, how, actually, just a few weeks period, how changes in air pollution can actually impact on health care. So integrating that and also integrating different types of data. So I talked about the wearables, but the other type of data that the citizen scientists collected was qualitative data. So they collected photo, audio, video that contextualized the numbers. So you could actually say, OK, why is it high there? OK, that's a busy road, there's construction there. And you can actually, it brings in the lived environment component, too. So I think it's really critical that we're looking at mixed methods, qualitative and quantitative integration. Fantastic point, Joanna. I'd love to get your take on this. Yeah, I'm going to riff on a different sort of vector here. During the COVID crisis, the very early days, we got the sequence of the virus, and that was really helpful for data science and some solutions. But what we realized very early on is there was no viral ontology available. Like no one had really done the work to map out viruses and the interaction of various medicines with viruses. And there was no way to train the AI models to look for potentially other antivirals that might be effective. It was very strange, like dynamic. I was like, I couldn't believe it. We were looking everywhere, there was nothing. So we looked for existing drugs, but not in the context of the virus. We looked in the context of other existing drugs that might be useful to combat COVID-19. And we found just by luck that there was an antiviral effect of a rheumatoid arthritis drug that meant that it would be an antiviral. And we proposed this and published it in the Lancet. And the company on the drug literally said, you people are crazy, you're going to ruin your company. But there's no way that there's a soft target effect in this drug. But we saw that signal in the data using our models, just running across all the available medicines. But it was just striking to me that there was no infrastructure in place. I mean, how many pandemics do we have to have before we have an infrastructure in place? We have all this technology. But to make a long story short, this discovery led to a drug that, for hospitalized patients, reduced mortality 38% on oxygenated patients. It was 46%. So if that information had been available earlier on, I mean, how many people could have been safe? So finally, we got full FDA approval, like Q4 of last year. That took a long time. So even when you come up with something that is provable scientifically, you still have to go through the trials and everything else. I just feel like we need to learn the lessons of that. And we need to look at this nexus between climate and public health. And what do we need to do? Like, I was reading the report in the Lancet about dengue, for instance, and the increase in prevalence of dengue going up. I think it was 12% increase in airborne transmission due to climate issues. We're working with drugs for neglected diseases here in Switzerland on drugs that can be repurposed for dengue because it's a disease that just has no treatment at the moment. So we need to mobilize, but we need to learn from what happened to all of us in the past couple of years. We can't just go flat footed into the next crisis, or poly crisis, as you say. Just quickly, staying on this theme of dengue is a great specific example. Do you feel like we have the data? It exists. It just hasn't been put together so that we can use it against machine learning models and be able to, with some accuracy, say, this is where it's going. These are the places that need to be on alert. We need to preposition medicine. We need to train health care workers. Is that the case, or is it more? Actually, we don't even have the data today. I think we have the data. I think if we, I mean, the challenge is normalizing across no one collects data in the same way. It's very difficult to make sense of it. One thing we've done, and this is, you know, you guys were all talking about large language models. And one thing we've done really well over the last four and a half years since I've been leading the business is run machine learning models across all the world's peer-reviewed research and mine that data for insights that might be useful for scientists. Now, the big thing for us now is how do we take this, you know, this vast corpus of knowledge across not just peer-reviewed papers, but we use like 85 other different data sources that we normalize across. And how do we get the power of those models into the hands of tens of thousands of scientists? Because I do think the data is there. I just think it's about mobilizing and getting... So what we do is try to illuminate disease biology, give scientists, I say superpowers, to see into what's happening in a disease and understand the network around it and how there are other sort of novel targets you can modulate with a drug that aren't being thought of and to come up with new ideas because as we all know, like drugs don't work that well for the patients who take them. They're like a blunt instrument. And, you know, especially for complex multifactorial diseases, I mean, we're making a lot of progress on gene therapy and, you know, there's a lot of cell therapies and things that are really quite precise, but for a lot of multifactorial diseases like TakeNash or something which is, you know, a lot of it is environmental, by the way. These are diseases that people have that are really complicated, diabetes, you know, and there's not great solutions today. Carl, when you think about the space, you're trying to kind of create the climate health nexus space. Are you aligned with Joanne on this? Do you feel like you're not gonna go out there and fund data collection? You're gonna fund more coalitions to take existing data, put it against existing problems using AI tools? Is that where we are at this point? Well, so I loved the example you used, Dengue, and I'm looking at my friends over at Welcome Trust because I'm sure they lit up when you said it as well, because Welcome Trust and Rockefeller and others have been working specifically on Dengue using existing data, building the models, building the tools to then do the predictions of where it's going to happen at this point, pretty high degree of confidence. And, you know, we're doing this across Brazil, Vietnam, you know, and building the evidence base, right? The randomized controlled trial that Welcome Trust is doing in Vietnam on Dengue is a great example of you've gotta show that it works, and then when you show that it works, then people will be picking it up, they'll be using it, because at the end of the day, that's what it's about. It's no longer about collect the data or analyze the data, it's about giving it to the people who need it and then supporting them in acting upon it. And so I think this was a great insight that you shared, which is that it's about the end user. This is another thing we learned from tech, right? It's the end user design that matters. You know, like if you hand somebody an iPhone and they wear it like a hat, they're not wrong, right? The user is never wrong, and that's a difficult thing when you're doing product development because users will do crazy stuff with your prototype, but you have to be really humble in product development and you have to be really humble in human development as well. I wanna bring in the audience, and we have a lot of people in the room, a lot of expertise on this to tell us what we're missing. If anyone already has a burning question, feel free to raise your hand. Otherwise we're gonna keep going a little bit while you think of one. It seems like there is a little bit of a tension here, though, between the way you're just describing this, and what we're hearing from Carl and Joanna, and the tension that if I'm hearing it right, is there's an idea of liberating the data, but you wanna make sure, even from the data collection standpoint, it's originally being collected with the end users in mind, and they're part of that process. Yeah, because data is not just data. You know, who informs, what informs what sensor you used, who decided where it was going to be placed. So what intelligence did you use to say, okay, we're going to measure this part versus that part? That's not scientific knowledge. That's kind of lived environment knowledge, right? So if you're not engaging with people, and you design something really fancy that is scientific sense, you have to engage them from the start because getting a sense of knowing what you don't know, like on the other thing, what you may not know because you're not in that environment is really critical. I wanted to pick up, if I may, on a point that you raised, which I really like the example you use of COVID, actually, something different, but what you were talking about in terms of the infrastructure that we actually have, reminded me of the importance of having this in place for early warning systems as well. So even for preparedness and early, say, outbreak, early notification of outbreak. So an example in the UK was that app that we were all using, when it first started to say, okay, report whether or not you have symptoms, right? And it had so many people using it that you were able to actually make sense of and spot patterns before they were actually confirmed. The only reason that was possible is because it existed before the pandemic hit, right? So it was something that was in place that was repurposed for COVID. So I think for me, it really underscores the importance of having that agile, you used the word agile earlier, agile data infrastructure that you have in place for whatever that you know that you can pivot to say, okay, this something, we having an indication of something or other, we can pivot it and say, tell us about this and get early indications from a general population perspective. The same applies by the way in terms of preparedness for the healthcare response, right? So we haven't talked about healthcare when we're talking about health and climate here, but really critical for that. If you get early indications of an outbreak coming up, what implication does that have on the components of your health system, of your personnel, getting the beds ready, getting the products ready, because sometimes that procurement takes time. And you can get those and again, and we have those data in that people know, but we're not asking people, we're not seeing people as important sources of data, but I don't like how I just phrased it because that sounds very transactional, but if you talk about citizen science, I like to think of it as redefining what it means to be a scientist, as well as what it means to do science or contribute to science, as well as what it means to be a citizen, right? So there's a civic engagement. Can you imagine a different world where being a citizen is not just you vote every four years and you do that, but actually a part of it is, I participate in this data infrastructure where I tell you about my quality, I tell you about my water quality, I tell you when I have a cold, I tell you. And this is part of being an active citizen. This is our power and I think this way because we have a demographic dividend on the African continent that we are not harnessing at all. You wanna make a quick point on this? I was gonna ask someone else, go ahead. I love Davos for these types of conversations, but I hope you won't mind me saying this, but I spoke yesterday to the Minister of the Economy and the UAE and they have a cooperative economy bill where if people contribute data to the government, like in a health environment, they get a piece of that. I think it's brilliant because like what you're saying is these are human beings and it's much more likely that people participate if they get a share in the company that, do you know what I mean? It was like a brilliant idea. I mean, I've heard it tried, there's agriculture cooperatives and there's cooperatives historically, but applying it in data context, I thought was absolutely brilliant. I hope he doesn't mind me sharing that. I think it's a great example, so probably not. I see a couple of hands are up. I'm just gonna ask one really quick one and then I'm gonna go to Jonathan here and see if there's any others. So you're talking about the implications, for example, for healthcare workers and health system. There's a whole global health infrastructure. Many billions of dollars spent every year by the richest donor countries, through bilateral aid agencies, through the World Bank, the WHO. To what extent do you feel like the existing global health infrastructure that's out there right now supporting public health systems is tuned into this discussion? Are they aware, do they think about climate and health, do they see the ground shifting in front of them? Are they thinking about the data they're collecting? Just a quick sense from all of you and then we're gonna jump to questions. Starting with you, Carl. Yes and no. So there are a lot of parts of the system and there are a lot of people in the system that are absolutely part of this. Yes. There are also a lot of the parts of the system that need some fundamental change and improvement. And there are some people who have not yet gotten to this discussion and I don't think for any malice or anything but I call it the day job problem. Everybody has a day job. It takes up your full day. Adding another sort of Davos sandwich on top of that is really tough. And so I think that people's intentions are in the right place and there's a lot of leaders in the space. And we're so excited to be working with several of them on a lot of these really important topics. So I'm hopeful we'll get the improvement that we need. Okay. I think I'm gonna go to quickly to questions then. I see Jonathan here, if you can bring him a mic and anybody else wanna, one in the back and one in the front. Okay, we'll take a couple at once. Go ahead. Great discussion and I heard Raj summarize his tables conversation in a different panel and you apologize for not getting the participants to get into a fight. So I like the attempt to just try my best, Jonathan, try my best. My question and it was joined a little bit to the point you were bringing up with the UAE but one of the things that we struggle a lot with my organization equips community healthcare workers so they can be responsive to core health system strengthening and then we argue pandemic preparedness. We had 100 plus thousand community healthcare workers equipped pre COVID, COVID hits. We got money from Rockefeller, many organizations to go do COVID-19 response. Zero of those community healthcare workers were able to get an updated application not because our software couldn't do it but because the government can't reprogram 50,000 workers jobs overnight. They couldn't even do it over a year though. So we stood up parallel workforces in both high income and low income settings rather than using the current ones. So at that point on the data collaborative or climate analytics or citizen engagement, these are horizontal skills we need today, tomorrow in the future for climate, for health, for all these things but the ROI of that public-private partnership is nearly impossible to articulate, and sell politicians on doing. So I'm curious as you think about this in all three of your respective roles, what can we be doing to somehow get that ROI? I think everybody hypothetically knows the ROI but it's a night job ROI, it's not a day job ROI. So how do we bring that in such that these collaboratives can really take root and you get a 10x return because it's not gonna end up just being that the climate data is useful and now you have a traffic sensor and you have all this stuff that you can keep compounding but it feels like we can't get over the hump of starting that process. We bring the mic over here to the front, we'll take a few at once. Jane here, I think you had your hand up. Thank you, yeah. So Jane Burson from the Clean Air Fund. I wanted to ask Tolu a question because I was struck by something that you've told me in the past about how African students are really held back from doing the analysis and research part of their PhD because they spend 95% of the time collecting the data that they then need to do their PhD. And I wondered if any of you have heard of ways that especially with health systems, they have fast-tracked the collection of data or people have got around the issue in other ways. Great question. And if you can pass the mic back to the woman, two rows behind you. Good afternoon, my name is Bata Gilisitze and I'm a global shipper from Khabaronehab in Botswana. So I'm gonna ask more from, I think they say from the, I don't wanna say from bottom, but from more community angle. You talk about some of the needs that are identified. As you know, Tolu, you'll know that a lot of the stories, a lot of the data that's available in Botswana on the continent of Africa is qualitative data. So in a community where the data is available in the form of stories, how can they package that in a way where they are proactive to say, this is what's happening in the climate. This is what's happening to our health. We can't come and give you the numbers and so forth, but we do need this kind of response. How can we package those kind of stories and that kind of qualitative data for a response that fits those communities? Can you just pass the microphone back behind you? We'll just take this last one. Go ahead. Thank you very much. I'm Enrique Pobianco from the WHO Foundation. Thanks very much for this incredible panel. I have one fourth dimension that I wanted to add to climate health and data and its equity. A lot of communities that are severely affected by climate are the communities where data is very difficult to collect and where digital innovations are very difficult to bring in. So I just wanted to add this dimension and ask you, what is the direction of travel to make sure that we really move in this area with equity firm in our minds? Thank you. Thank you for that. So I think there's two questions here that to me seem like leapfrogging opportunities over the problem. So when we start there, maybe with you Joanna. Jonathan's about CHWs. They seem like a huge opportunity. There's hundreds of thousands of millions around the world. And if they could be equipped with a smartphone connected to the cloud and if it had tools for data collection and sensing and if you could bring AI to it, you could do so much instead of saying, well, you're actually tasked with this project for this particular disease area. And then similarly on this point about the places that are toughest that hit the hardest, least likely to have the tools or the connectivity. So how do you think about these leapfrogging opportunities? Is there a way we don't have to rebuild everything in the hardest hit places? Is there a way we don't have to build a whole new infrastructure but use the CHWs that are there now? I do go back to this model national or model city response idea and you know there's, I just think the best way to orchestrate and roll out these types of solutions is for everyone to be exposed to best practice and best ideas. And I think the only way you do that is somebody in a multi-stakeholder organization has to collect it, set up the infrastructure for sharing it so that there's an immediate transfer of knowledge to everyone simultaneously. You know, and I really feel like it's not that hard to set up, right? And you can change it all the time. As new things come on as evolutions in applications, various other things, it just needs to be a global GitHub for solutions. You're exactly right. I hadn't thought of that, that was a really good idea. But if you can put together just action on specific problems, whether it's air quality, I mean, there's so many different groupings of this, but experts can identify and prioritize these groups and then invite solutions. This is what I measured at the beginning. It's the question of what level to enter the problem, right? Do you enter it at the city with dengue layer or do you do it at a much higher level, the Davos kind of level, that global infrastructure, the global healthcare system? Do you think about CHWs as a point of entry? How do we enter that problem? Carl, maybe you can bring that up, especially again, keeping in mind, Honorary's point that if we just look at where the infrastructure is today, we might actually make the equity problem worse. Yeah, that's absolutely right on that equity point. And there's a couple of different threads here, right? But I think it does come down to this combination between global and local, right? And to be fair, this is something we've talked about for decades. So it's not a new problem, but it is still a problem and it's being exacerbated by the other problems that are out there. You know, in terms of how we handle it, the unfortunate answer is we need to do both. The question is how do we do that? And I think the answer is that we do, we show up at places like Davos, don't be too cool for Davos, use it for good. You show up at places like COP, you show up at places like WHA, and there's a bunch of different places that we should be showing up, we should be having these conversations, but it can't stop there, right? When we have these conversations, we then need to say, so what now what? What am I doing? What am I doing today? I'm not gonna put it on a to-do list because next week my day job is gonna come back and I won't get to my thousand Davos to-dos. So what can we do today to take that global piece where we're having this amazing conversation and action on it in a highly localized way? And that localized way, that's gonna look very different by definition for each of us, so I can't tell you what it is, but you'll figure out what it is. When you figure out what it is, you act on it. And I'll give you a hint, the way that you'll know what it is is by listening to the end user. So figure out who's your end user, what do they need, and then what's the next step that I can take today to help that end user with their need? Well, I guess to Joanna's earlier point, if we find that real world problem and actually solve it somewhere, anywhere, it becomes the kind of thing that people start to pay attention to and say, well, how do I do that where I live? That's the story that takes on its own life. Everybody likes to win. And so if you can create a win, people just flock to it. Totally, just thinking about a couple of questions posed to you. The story in part, maybe we can start there with our colleague, the young global shaper from Botswana, she's saying, a lot of times communities know what's happening. Maybe they see the riverbed is drying up or we're getting more flooding. We don't have data, but we know what's going on. How can we better package and utilize that information? Okay, I'm going to answer that question. I'm going to take a step back and talk and answer and point to the leapfrogging point that you made. Just do it briefly. Very briefly, she says. Leapfrogging, often we're so focused on leapfrogging that we forget what we can learn from the past. I just want to highlight an anecdote in the 1940s in rural South Africa, there was an experiment called the Pallela experiment was in Pallela, rural South Africa, where they piloted what is essentially community health workers, but you had people who were responsible for households and it wasn't just the health, you knew when they had a baby, you knew when the mother lost her job, you knew when, so you picked up. You knew that, okay, you knew when their crops failed and if they subsistence, that's an early warning of potential stunting, right? And then you step in rather than just waiting in the healthcare for it to occur. So they were embedded in that way. Now, if you can imagine that model with the tools that we have now, we actually, we can learn from the past and we can superimpose the tools that we have we didn't have then. So coming up to your question, I think that's really critical. What you just described is entirely why some of the citizen science we do is set up that way. So we look to integrate quantitative and qualitative data because I don't, often we talk about, we make data synonymous with quantitative, right? You can use stories and with our citizen science, we talk about generating data stories, right? Because there's the numbers, but then you can also do things like geolocating the story, right? Which allows you to be able to integrate it in ways that you weren't able to before. So I think we can play around with platforms in that way. And then the last point around the students absolutely critical. My PhD students in South Africa spend two and a half years collecting data and then, I mean, that's not PhD knowledge, right? My PhD students in Cambridge is just like, oh, which data set do you want to use? And you spend most of your time designing a project, data analyzing data. You both come up with PhDs at the end, but your skill set and your ability to where you jump into the pipeline, very different. And so it's really critical and that comes to the equity question. It's really critical when we're designing this data infrastructure we're thinking about. Who has access to those data? There's some data on African cities that African students can't access because they were designed by and they're collected by and they sit in, right? So we have to think about, I don't need to finish my sentences because of time. So really critical when we think about how we design the, how we collect data, we have to say, again, end user, if a key part of our end user is training a pipeline of the next generation of data scientists, of researchers that are able to grapple with big data, with AI, then we have to ask, do master students, does a master student in Bujumbura have access to big data? About Bujumbura or about any African city or any African country? Because if not, we need to be addressing that. It sounds like maybe the beginning of a set of principles that Welcome and Rockefeller could push out there on data collection at this nexus. Maybe that's for the next of us panel. We're nearly out of time. I just want to give Joanna and Carl a quick last chance to wrap up anything on the questions totally just addressed or anything else you think we should be taking away from this discussion. What I'm taking away is that Raj just gave me some homework so I'll go find Welcome and we'll chat. No, but really, that is how I feel. I feel that there's a lot of good energy, there's a lot of good ideas, and this actually just syncs up with, yesterday I was thinking about this as well, is what do we need to build out that we have not already built, because some of it's built, what do we need to build out to be that sort of platform for good within the climate health space and other spaces? Like what is missing? And that's the homework that I've taken away. And Joanna, last thoughts from you? I'll just leave you with the, I mean, I'm an internal optimist, silver lining person, and I look at the potential for increasing productivity and creativity of large language models. I know there's a lot of dangers and I know that they hallucinate and they predict really crazy stuff and that's crazy, but we're going to work that out and we're going to turn this into something immensely valuable for society. Like just the summation capabilities of bringing democratizing access to information to everyone is so powerful. I've been in digital since 1985. I've never been more optimistic about the opportunities and to finally deal with this equity problem because there really is no barrier, but as long as we train these things with the right data that's not biased and we have to be stewards of that, but the potential is enormous. And there's really no other way, right? Given how we started the discussion, given the immediacy and the trajectory, if we don't leverage these digital and data tools, there's no other way we're going to get ahead of this set of issues. So that gives us some hope as we close this panel. I thank all of you. I thank Welcome and Rockefeller for the pleasure and please join me in thanking our great panelists.