 I think that the AI for Good Summit is really the first time that we've seen industry, the technology industry and people with a lot of sense of human needs coming together to look at the future of this particular technology, the space around machine learning and more developed neural networks. I think that as an output from this, we hope to see what we've seen in other fields and over the last few years UNICEF has been working closely with the industry around drones and UAVs and what we've seen is a consolidation of players but also a focus of effort. What we find is that if we can help to define some of the greatest needs in the world and also a path to profit, how business can be involved with those needs whether it's 50 million refugee children who are on the move because of violence and how businesses can work with that community to create stronger businesses and help humanity that we can create both more consistent and solid businesses but also help those who most need it and I think that we hope to see from this conference is really that kind of collaboration emerge. You have about 500 million kids who are affected by food insecurity either they don't have enough food because of too much water or because of not enough water and it's due to climate change and the devastating impacts that that's having on farmers and that's a need on the part of children if they don't get the right nutrition they don't grow their physical body doesn't grow their mind doesn't grow and that means they can't be full producing members of society. We think we can work with people in the space of machine learning to try to understand that problem better to figure out where schools are where we could provide feeding programs where the most vulnerable communities are where we could try to input something and that would create both a stronger community but also really a stronger world. I think the discussion today has been fascinating. One of the things that I see as a very interesting problem to tease out over the next few days is a disconnect and it's a disconnect that sort of falls along the following lines. There are a bunch of people with great tech skills and great networks who are able to produce technology that helps other people who are wealthy and well connected. And then there are a bunch of people who aren't included in the conversation at all either because they're too poor to be here in Geneva or because they don't have access to the networks that we have. And what I see as a really interesting problem to think about over the next few days is how we try to take that divide which is going to grow worse and worse over the coming years and try to make it connect a little bit more. And for me that means doing things like looking at you know we heard a lot this morning about genetics and how AI can be applied to the space of genetic health for example and personalized health care. But what does that mean if you don't have the genome for half of the world's population and you're trying to create genetic based genetic medicines that that help heal everybody. What it means is that we're going to have more disparity rather than less and more disparity leads to more inequality and that's really bad for everybody. There are a huge number of kids who don't have the right nutrition and and if you don't have the good calories that you need your body becomes stunted your brain becomes stunted and the only way right now to see that is to look at your upper arm circumference and your height and your weight and it's really hard to weigh like a four-year-old kid because they're wiggling all over the place. And if you can combine those three indicators you can figure out whether the kid is very malnourished moderately or he's or she is very healthy. But we also think you could use image recognition and facial recognition technology to do the same thing. Now that's not a very sophisticated use case and if you tell a lot of the people in the AI field if this is something we want to do they'll sort of say well it's okay to solve problem but that problem could directly impact half a billion kids and so you know how do we put the pressure on to make sure that those simple problems which might be well described and kind of boring are solved first and I think that's a challenge that we have in the next two days. For UNICEF that means that we have to look at our data as a real source of value that we have offices in 190 countries around the world and we're able to get data ground truth data training data that companies want in order to make their intelligences more intelligent and if we know how to value that data we can actually come to the table not with a begging bowl but with a joint venture and so that's a kind of neat way of positioning it. We've just joined the partnership for AI for good as one of the founding members and that gives us a position where we can sit with industry leaders and speak together and I think that's really important too. We're not trying to chase a train that's already left the station. We're trying to have kind of a train that we build together. I see that AI or machine learning can be most helpful at is looking at the core underlying issues. Why is it that people don't have access to water? Is it that they're too far from a city from a water point? Is it that they don't have enough money? Is it that they don't have enough of something else and we as humans can't look at that data and understand it so we find like one thing one answer and we kind of work on that or a few and our water people are amazing and they're brilliant but we've never had access to this much data before and so now we can look at all the data that we can get and apply complex machine learning techniques to it and have some intuitions come out that we can test and so that's a very cool opportunity in a field that I think has some of these intractable problems and water and access to water is going to be one of the fundamental things that divides the wealthy from the poor in the coming years and you can already see the the edges of that problem show itself but underlying that is a lack of fair access to opportunity and choice.