 We started the Science for Social Good program in the summer of 2016 and the idea was to kind of tap into our scientific talent to try to find novel technological solutions, novel solutions grounded in data and artificial intelligence that could be used to address most important social and humanitarian challenges of the world. The culture of AI research and development needs to be fully intertwined with social good problems. The idea is to partner with NGOs, public sector agencies, social enterprises to understand the problems that they're facing because they are the people who are at the front lines of the challenges. Over the last three or so years, we had over 100 scientists contribute their skills and expertise to our projects and I think that's really something really unique to IBM research because I can think of very few places in the world that can offer that kind of breadth and scale of expertise and passions. We've created a lot of case studies of how AI can be used for advancing social good. One of the things that we are trying to do with this program is to try to create really scalable technologies, technologies that can be reused across many problems. If AI is improving with the problems that we're facing for the world, then really it's humanity that's getting uplifted at the end and that's what we want. We become better scientists and better engineers and better human beings and I think that's probably the main point of running IBM Science for Social Good.