 So one of the things that I get to do and have gotten to do over the last ten years is to visit some pretty incredible places where amazing work is done. And about seven or eight years ago I got to go to Boulder to be a mentor at the Unreasonable Institute with this particular class of entrepreneurs who were there, and I got to meet an amazing man, Luis Duarte. Luis at the time was working in recycling garbage in Mexico and being someone who can pivot and move on, he is now head of the Geary Community Investments, focusing on early childhood education. And I had a great moment backstage when I got to say to him, welcome my green card friend and had everyone look at me like what kind of political incorrectness did she just step into. But I just want to tell you, I want to welcome Luis Duarte, my green card friend to the stage and he'll tell you why. Thank you, Rosalie. And I think first of all, it's really a pleasure. I'm not the head of Geary Community Investment. We have a ton of beautiful colleagues that are representing our organization, but I do love coming to SOCAP. This is home. The first SOCAP that I came in thanks to Tony Carr was in 2012. And I have been since introduced to Geary Community Investments thanks to Tyler Hartman and Jed Emerson. And Geary Community Investments, which includes the Python Foundation, invests in for-profit and philanthropic opportunities to help advance the lives of low-income children and their families in the state of Colorado. So last year, a group of us that were here as we were heading back home in Denver, we wondered about this question and the question is how might we leverage this community? How might we leverage SOCAP and show each of the participants, each of you, the correlation between early childhood, social determinants of health, economic development, neighborhood economics, and also sustainable livelihoods. And also share with you the incredible financial and social return on investment when you're looking at this field to make a commitment. Thanks to Rosalie, Kevin Jones, Lindsey Smelling, and Kerry Hanson, Geary Community Investments has been working this whole year to curate the first early childhood spotlight series. We're very excited to bring you not only our hashtag, SOCAP4Kids, but also to announce a one million early childhood innovation price to really unearth solutions that will help improve the lives of children zero to three. And what better way to start this spotlight series than with our next speaker? I'm truly honored to introduce and welcome to this home, to this community Dr. Satya Nita. Dr. Nita is a worldwide leader and program director of the Cognitive Sciences and Education Technology and Research Department at IBM Watson. His global team invests and develops technologies at the intersection of cognitive neuroscience and cognitive computing and employs multiple techniques in fields that are ranging from machine learning to natural process, natural language processing, virtual and augmented reality to cognitive neuroscience. He has been the chief technologist behind several critical innovations in the silicon technology area. Dr. Nita has been honored with multiple awards, including the IEEE Spectrum Innovator of the Year. And in 2016, he was chosen as one of the top 50 worldwide innovators, movers and shakers of education by the World Innovation Summit. Aside of his incredible credentials, I was so touched, inspired and humbled to hear his personal journey. Please join me in welcoming Satya. Good morning. It's a pleasure and honor to be here and I'll blame Louise for spending the first two or three minutes of my talk, giving you a little bit of a background behind my journey into early childhood. So there's something called the the Star Trek phenomena. It's actually very well known amongst technologists, which is that at a very early age, a lot of children get exposed to TV shows like Star Trek or to science fiction and they're so inspired by it, they grow up dreaming of building this world that they see. So I was very much one of those people. I spent a good part of my teenage years reading nothing but science fiction and sinking in as much astronomy and astrophysics as I could and and that translated to spending most of my professional life working in nano-electronics and nanomaterials and working on silicon technologies. So how did I end up in early childhood? So the story is, you know, IBM takes people like me and gives us all kinds of exposures and one year they asked me to spend a year shadowing the one of the senior leaders of the company because they wanted me to learn how to run big organizations and manage large budgets. So I spent a year doing that and at the end of it they said what do you want to do next and so I'd spent the last several years with with a real fascination for a branch of computing called neuromorphic computing and which is this notion that digital computers look very different from the human brain and there is a reason for the difference and how do we bridge this gap. So I got very interested in that field and realized I wanted to spend some time pursuing the software side of things and got into AI. So when I landed in AI, I had to find an application area to work in and Watson had just one jeopardy and about a year later they were looking for somebody to start a team trying to figure out what to do with Watson in education. So I was inspired by the fact that there was a big idea behind this whole thing just like there was a big idea behind the the technologies that I spent doing my earlier work in and I decided to take up the the challenge and the opportunity. So the big idea here is that you know computing is evolving rapidly a lot of young children are exposed to experiences that we've never seen and and I was fascinated by what things like touch computing or interaction with a virtual agent will do to how my children and this generation will learn and how their brains will be wired and how they'll be different from us and think very differently from us. So I loved that big idea and I said okay, I do want to spend the rest of my time in AI doing early childhood education. So back here is actually an image of Saturn. This was taken by the Cassini spacecraft three days before it actually plunged into Saturn and so this ideas like this I mean images like this still inspired me. I showed this to my children. They were captivated by it. So it made sense to me that you know bridging big ideas and technology and trying to transform early childhood education was where I should spend the rest of my time. But what really changed things for me was a talk that I saw by Patricia Kool. It was an inspirational talk. She's a very famous researcher in this area and she had a lot of things to say about brain development and growth. But one of the things that she said that stayed with me is the rapidity of brain growth in the first five years of life and the investment in public education that happens. So right at the point where the brain is going through the most important and dramatic period of growth, we have very very little investment in public spend and education and this is a real pity because everything we see with respect to people dropping out of schools, colleges, having poor life experiences can be traced down to this exact period where there is neglect, there's poor nutrition, very few learning experiences. And so this caught me really really interested in the problem. And then as I was reading the literature, I came across this paper from John Gabriel Lee's group at MIT where he talks about the actual physical impact of neglect in the early childhood timeframe on children's brains and he talks about the difference in the thickness of the cortical sheet in the neocortex depending on, you know, which, what kind of background you come from. And this absolutely stunned me. So there is an actual physical change in the brain and a physical lack of brain growth that you can measure through FMRI in early childhood. So this caught us thinking a little bit about what can we do with technology? Can technology help level the playing field? Can we do something to actually address this gap? And we sought out a partnership with Sesame Workshop, who's of course well known for a tremendous amount of work in early childhood and delighting generations of children and teaching them words and concepts and so on. So Sesame was delighted to work with us and so when we formed the partnership, we had a couple of principles. The first was we realized that despite IBM and Sesame being very big names, we realized that we're just a drop in the bucket and to make a difference in such a big field across the globe, we need an entire community of people to come together. So we said, you know, one of the things we ought to do just by partnering together is to draw attention to this field and channel spotlight on this field. And the second thing we ought to do is to figure out how to use AI to transform this particular field. And it's a very interesting time in AI, as all of you know. The notion that I can dialogue with a computer, that a computer can understand natural language, that a computer can understand speech and vision, and take some decisions and reason. All of these things are at the frontiers of AI today and which is what we were doing in our day-to-day life. So we thought partnering together and using these capabilities and doing something that will first of all shine a light on the problem and second enable a lot of people in the field to build their own technologies would be the right thing to do. So what we decided to do was to build a platform to impact early childhood learning and this platform would be basically used to create a generation of smart apps and toys and games and so on. And so this is a software platform. It's still in development, but we decided that the platform was basically the way to enable an entire ecosystem of people who don't otherwise have the means and the mechanisms to use AI to create their own technologies and impact the space. So I want to talk briefly about what the platform does. So first of all we started with the core Watson technology itself and Watson technologies today involved things like speech and image recognition, conversation between a computer and a human, the notion of many dynamic types of character voices. So these are all the base foundational elements of the platform and when we looked at it we realized that none of the technologies actually work for early childhood. First of all understanding children's speech is very difficult for a computer. It's a unique problem. Having meaningful conversations with children is very different from chatbot answering questions for a call center. So we spent a fair bit of time in the first year of our partnership tweaking the underlying Watson technologies and creating a more robust platform for early childhood. And then we said we can do something a little bit more ambitious and we started building a series of custom AI services as we call them for early childhood and which is the layer that's on top. So what we started building was a network of all the concepts in early childhood. We started feeding Watson all kinds of literature that children normally read. So we have an idea of the concepts, the vocabulary, the words that they have and and this is semi autonomously generating this entire knowledge graph of early childhood concepts and using that we could do some very interesting things. We could we could have a computer generate assessments. We could have a computer generate personalization models and and so this platform then was basically the foundation by which we could start building technologies ourselves just to test it out. So the first technology that we built was a learning application. This is a vocabulary learning application to address the the well-known 30 million word gap that all the people in this community really understand. And so we started piloting this. We piloted this at a very large school district in Georgia at Gwinnett County Public Schools and and this slide you see the the relationship is a little bit better. So based on that we started getting some real really interesting data. For instance the the graph at the bottom left is basically the the vocabulary development for one particular child over a 15-day 10-15-day period of being exposed to the the learning experiences that are built within this app. And the larger circles represent more confidence in the word, smaller circles represent less confidence in the word, and you can actually see a kindergarten-age child actually grasping words as they go through and use this particular application. Among other things the application can also generate assessments based on the the knowledge graph that you see below. It will ask a question like which of the following mammals is capable of flying and it will also generate automatically all kinds of distractors and it will have this very conversational interface with children. We mean this application to be an exemplar application of the platform. We don't intend for this to be the only application that children will use. And so among other things we also thought it would be interesting to use the platform to power a generation of smart toys which can have very interesting conversations with children and among other things they can also situate the conversations within their surroundings. For instance the toys might be able to look at a child's surroundings and they might be able to have a conversation based on the characters that they see or the objects or the toys that they see. And all of these are enabled by computer vision and speech recognition and the network of concepts that we're building. So these are meant to be some example applications of what one could build with the platform but what we really intended was over the next year or so to open this up to the world and let people build their own learning applications and tackle major problems like dyslexia which is a possible future application and unleash kind of the innovative power of an entire community. Dyslexia is a great example of what one could do with a platform like this that's very well known today that dyslexia is actually really a problem in the phonological processing areas of the brain and the ability to distinguish sounds is also very critical to the ability to read. So you could use speech recognition which is built on understanding phonemes in very clever ways to address problems like dyslexia. So I want to end with a little bit of a call to action. I'm very humbled to be here very honored to be here amongst a network of you know influences, finances, etc. I would like all of you to consider investing in this area and consider you know investing in technologies and people who can scale the problem, scale their reach and allow many more children on the globe the opportunity to grow up and dream of stars and reach for the stars. Thank you so much.