 We all know that AI is transforming the workplace but it's starting to transform our personal lives too and entering our homes. This matters. The home is a special place. It's where we care and nurture for the people we love. It's where we endeavor to flourish. It's where we want to raise our children for a better life and age with dignity and independence. But unfortunately we're living in a time where the pursuit of happiness is becoming increasingly difficult for many. Chronic loneliness and depression are on the rise and a silver tsunami of a global aging society is causing an ever-expanding care gap where now the projected number of professionals and institutions can't scale to meet the growing demand. So what if we could design AI that not only helped us to live more productive lives but happier lives too? Can we design AI to support a deeper human experience to help us unlock our human potential so that we can become who we aspire to be? Can we design AI to promote human flourishing? This would be quite a different AI from what we designed today. We would need to have these systems become socially and emotionally intelligent, especially supposed to support us in areas of personal growth. But wait, you may say, does this mean we want machines to replace us even in areas of care and support? And the answer is, of course not. We need to respect our fundamental human need to belong and to be valued by others. However, it is a question of social justice if we risk a feature where far too many will go without quality care and support, that's affordable. So AI, it can be designed to bridge the demand. It can scale in an affordable way. And importantly, it can also be designed to be human-centered. For instance, my group at MIT, we designed a technology called social robots. Our goal is to create socially and emotionally intelligent machines that are natural for anyone to be able to collaborate and interact with. So rather than leveraging autonomy to design machines that can perform tasks independently from people, social robotics is about designing robots to be teammates that work with people to help us achieve our goals. To do this, we have to design these robots such that their behavior naturally dovetails with how we as people think, learn, behave, and collaborate with others. This requires an ever-increasing depth of understanding of human intelligence and human behavior. This is not to replicate or replace human intelligence, but rather to better understand how to design AI so that it works with us. Over the years, we've designed robots that help people to succeed, thrive, and grow. From emotionally aware tutors to help children learn new skills, to helpful companions, to help seniors age with independence, to even personal health coaches for adults. What we're discovering is that robots can actually engage human psychology and behavior in very deep ways. It's almost as if people can form an emotional bond with these kinds of technologies, that they can form a positive human-robot relationship. It is almost like having a really smart pet that can actually help you achieve your personal goals. Machines are starting to be able to interpret human emotion with greater nuance. We're finding that the design of the machine's own emotional behavior is critical for building this trust and this positive rapport. Personalization also matters. Over time, robots are starting to be able to learn from interacting with us how to adapt their behavior to help support us in the most effective, efficient way. This social and emotional dimension of our relationship with machines is this new territory for the human-machine relationship. Just as our friends and our allies can shape our attitudes and behaviors, we're discovering that robots can be socially persuasive as well. This raises important ethical considerations as now we can design AI systems that can socially and emotionally support us to achieve our goals or potentially manipulate us to benefit someone else. We're at an inflection point that's going to shape our future with society and AI. It's going to really depend, I think, on how we answer three questions. What is the nature of intelligence we want to design? Who is empowered to design and control it? How do we want it to be implied and to support what values? We can design machines to replace people or we can prioritize the design for a powerful partnership with people. If they're going to be able to collaborate with us, they will need to be socially and emotionally intelligent to dev-tail with us. It's also important for us to think about AI literacy for society. As long as only a privileged few have the training and education and power to control how AI is used, we risk a future where AI is going to accelerate and exacerbate the prosperity divide rather than to close it. We need to democratize AI. We need a far more inclusive, diverse group of people educated and able to control and use and apply AI in a responsible way. To date, so much of the conversation on AI has been dominated by topics such as efficiency, cost-effectiveness, productivity, but healthy societies and economies don't just grow, they thrive. So I ask you here as participants, how do you think AI should be applied in society? And what policies might you think need to be in place to help ensure that AI helps everyone to prosper and flourish? Thank you.