WEBVTT 00:00:00.000 --> 00:00:01.500 align:middle line:90% [MUSIC PLAYING] 00:00:01.500 --> 00:00:05.760 align:middle line:84% CHILD: Alexa, turn off the lights. 00:00:05.760 --> 00:00:06.675 align:middle line:90% AMAZON ALEXA: OK. 00:00:06.675 --> 00:00:10.890 align:middle line:84% CHILD: [GIGGLING] Alexa, turn on the lights. 00:00:10.890 --> 00:00:13.080 align:middle line:90% AMAZON ALEXA: OK. 00:00:13.080 --> 00:00:16.680 align:middle line:84% SARAH HANSEN: Today on Chalk Radio, robots and humans become 00:00:16.680 --> 00:00:18.845 align:middle line:90% collaborative partners. 00:00:18.845 --> 00:00:20.220 align:middle line:84% CYNTHIA BREAZEAL: It's not about, 00:00:20.220 --> 00:00:22.620 align:middle line:84% let's build machines that look like people 00:00:22.620 --> 00:00:23.580 align:middle line:90% and act like people. 00:00:23.580 --> 00:00:26.850 align:middle line:84% It's, let's build machines that dovetail with us in a way 00:00:26.850 --> 00:00:29.670 align:middle line:84% that we can bring forth all of ourselves 00:00:29.670 --> 00:00:31.440 align:middle line:84% to help us achieve the goals that 00:00:31.440 --> 00:00:33.750 align:middle line:90% are really important to us. 00:00:33.750 --> 00:00:36.300 align:middle line:84% SARAH HANSEN: I'm your host, Sarah Hansen. 00:00:36.300 --> 00:00:40.290 align:middle line:84% This week, we're talking with a researcher and educator who's 00:00:40.290 --> 00:00:43.230 align:middle line:84% working to bridge the gap between human well-being 00:00:43.230 --> 00:00:45.420 align:middle line:90% and artificial intelligence. 00:00:45.420 --> 00:00:48.300 align:middle line:84% From kindergarten to college classrooms, 00:00:48.300 --> 00:00:50.310 align:middle line:84% she's preparing the next generation 00:00:50.310 --> 00:00:55.245 align:middle line:84% of informed technology users and empathetic designers. 00:00:55.245 --> 00:00:56.870 align:middle line:84% CYNTHIA BREAZEAL: I'm Cynthia Breazeal. 00:00:56.870 --> 00:01:00.860 align:middle line:84% I am a professor at MIT at the Media Lab. 00:01:00.860 --> 00:01:05.630 align:middle line:84% I am also director of a new MIT-wide initiative 00:01:05.630 --> 00:01:09.380 align:middle line:84% called Responsible AI for Social Empowerment and Education. 00:01:09.380 --> 00:01:11.210 align:middle line:90% We call it RAISE for short. 00:01:11.210 --> 00:01:16.122 align:middle line:84% And I very recently am the Dean for Digital Learning. 00:01:16.122 --> 00:01:17.580 align:middle line:84% SARAH HANSEN: Cynthia's fascination 00:01:17.580 --> 00:01:20.550 align:middle line:84% with artificial intelligence began a long time 00:01:20.550 --> 00:01:24.298 align:middle line:90% ago in a galaxy far, far away. 00:01:24.298 --> 00:01:26.340 align:middle line:84% CYNTHIA BREAZEAL: When I was 10, I saw Star Wars. 00:01:26.340 --> 00:01:30.030 align:middle line:84% And that forever shaped my, I guess, 00:01:30.030 --> 00:01:32.937 align:middle line:84% vision of what a robot could be for us. 00:01:32.937 --> 00:01:34.770 align:middle line:84% I mean, it wasn't just a mindless automaton. 00:01:34.770 --> 00:01:37.035 align:middle line:90% It was an entity that had-- 00:01:37.035 --> 00:01:39.035 align:middle line:84% in the movie, they were full-fledged characters. 00:01:39.035 --> 00:01:39.535 align:middle line:90% Right? 00:01:39.535 --> 00:01:40.440 align:middle line:90% They had friendships. 00:01:40.440 --> 00:01:43.170 align:middle line:84% They were mission-driven to save the universe. 00:01:43.170 --> 00:01:44.490 align:middle line:90% I mean, all of that. 00:01:44.490 --> 00:01:46.652 align:middle line:84% And, of course, they had relationships, 00:01:46.652 --> 00:01:48.360 align:middle line:84% not just with the robots but with people. 00:01:48.360 --> 00:01:49.400 align:middle line:90% [MUSIC PLAYING] 00:01:49.400 --> 00:01:51.150 align:middle line:84% SARAH HANSEN: Cynthia followed that vision 00:01:51.150 --> 00:01:54.960 align:middle line:84% of complex human-machine relationships all the way 00:01:54.960 --> 00:01:57.810 align:middle line:90% to a PhD program at MIT. 00:01:57.810 --> 00:02:04.493 align:middle line:84% There she designed Kismet, the world's first social robot. 00:02:04.493 --> 00:02:06.910 align:middle line:84% CYNTHIA BREAZEAL: That kind of started this whole question 00:02:06.910 --> 00:02:09.310 align:middle line:84% of socially and emotionally intelligent interaction 00:02:09.310 --> 00:02:12.340 align:middle line:84% with machines, because we are profoundly socially 00:02:12.340 --> 00:02:13.780 align:middle line:90% and emotional creatures. 00:02:13.780 --> 00:02:15.250 align:middle line:84% If we want to design technologies 00:02:15.250 --> 00:02:17.930 align:middle line:84% that can interact with us and treat us, again, as people, 00:02:17.930 --> 00:02:20.650 align:middle line:84% we need to understand that dimension. 00:02:20.650 --> 00:02:22.570 align:middle line:84% A lot of the early work was pushing 00:02:22.570 --> 00:02:24.890 align:middle line:84% on I would say the algorithmic side of that. 00:02:24.890 --> 00:02:28.240 align:middle line:84% We looked at a lot of models of emotion, and people, 00:02:28.240 --> 00:02:30.620 align:middle line:84% and animals, and what is the function of it? 00:02:30.620 --> 00:02:32.110 align:middle line:84% And how might you implement aspects 00:02:32.110 --> 00:02:35.290 align:middle line:84% of that for intelligent communication and interaction 00:02:35.290 --> 00:02:36.560 align:middle line:90% with machines? 00:02:36.560 --> 00:02:38.440 align:middle line:84% And then we started thinking about, OK, 00:02:38.440 --> 00:02:39.732 align:middle line:90% so what's this really good for? 00:02:39.732 --> 00:02:42.080 align:middle line:84% So it's fascinating from a scientific perspective. 00:02:42.080 --> 00:02:44.260 align:middle line:84% It's fascinating from a technical perspective. 00:02:44.260 --> 00:02:46.900 align:middle line:84% But what does this mean about how 00:02:46.900 --> 00:02:51.640 align:middle line:84% we will interact with robots, with intelligent machines 00:02:51.640 --> 00:02:53.030 align:middle line:90% in our lives, in our future. 00:02:53.030 --> 00:02:54.640 align:middle line:84% And so then we started thinking about, 00:02:54.640 --> 00:02:59.140 align:middle line:84% so what are these big societal challenges that we face where 00:02:59.140 --> 00:03:02.710 align:middle line:84% potentially an AI cut more from this kind of cloth 00:03:02.710 --> 00:03:05.422 align:middle line:84% or this philosophy might really make a difference? 00:03:05.422 --> 00:03:06.880 align:middle line:84% SARAH HANSEN: After Kismet, Cynthia 00:03:06.880 --> 00:03:10.623 align:middle line:84% began working on a new social robot named Jibo. 00:03:10.623 --> 00:03:12.040 align:middle line:84% CYNTHIA BREAZEAL: Jibo is actually 00:03:12.040 --> 00:03:14.440 align:middle line:84% these concentric rotating circles 00:03:14.440 --> 00:03:16.060 align:middle line:90% that are kind of off-axis. 00:03:16.060 --> 00:03:18.910 align:middle line:84% And with that, he can kind of create 00:03:18.910 --> 00:03:20.810 align:middle line:84% what's called an animational line of action. 00:03:20.810 --> 00:03:21.880 align:middle line:90% So he can change his posture. 00:03:21.880 --> 00:03:22.380 align:middle line:90% Right? 00:03:22.380 --> 00:03:23.980 align:middle line:84% So he can look like he's upright. 00:03:23.980 --> 00:03:27.460 align:middle line:84% Or he can stoop over or lean to the side and cock his head. 00:03:27.460 --> 00:03:30.490 align:middle line:84% He can go through all of these kind of expressive postures 00:03:30.490 --> 00:03:33.790 align:middle line:84% that are really foundational to an animator when 00:03:33.790 --> 00:03:37.080 align:middle line:84% they want to express something through characters. 00:03:37.080 --> 00:03:37.820 align:middle line:90% [MUSIC PLAYING] 00:03:37.820 --> 00:03:42.430 align:middle line:84% Technologies tend to move in rectilinear trajectories. 00:03:42.430 --> 00:03:44.660 align:middle line:90% Living things move in arcs. 00:03:44.660 --> 00:03:47.660 align:middle line:84% And so all of Jibo's motions was based on arcs. 00:03:47.660 --> 00:03:48.160 align:middle line:90% Right? 00:03:48.160 --> 00:03:50.535 align:middle line:84% So he naturally moved in a way that just seemed much more 00:03:50.535 --> 00:03:52.630 align:middle line:84% lifelike and expressive, because he fundamentally 00:03:52.630 --> 00:03:54.940 align:middle line:90% was designed to move in arcs. 00:03:54.940 --> 00:03:56.357 align:middle line:90% You have to see Jibo move. 00:03:56.357 --> 00:03:57.940 align:middle line:84% You have to actually watch them dance. 00:03:57.940 --> 00:04:01.420 align:middle line:84% Because it's incredible to watch this robot dance. 00:04:01.420 --> 00:04:02.200 align:middle line:90% [MUSIC PLAYING] 00:04:02.200 --> 00:04:04.990 align:middle line:84% SARAH HANSEN: Jibo didn't just look and move differently 00:04:04.990 --> 00:04:06.340 align:middle line:90% from other robots. 00:04:06.340 --> 00:04:09.840 align:middle line:90% He functioned differently too. 00:04:09.840 --> 00:04:12.240 align:middle line:84% CYNTHIA BREAZEAL: I often use the language of there's 00:04:12.240 --> 00:04:14.070 align:middle line:90% the Digital Assistant. 00:04:14.070 --> 00:04:15.570 align:middle line:84% Like, this is what like Google and-- 00:04:15.570 --> 00:04:17.153 align:middle line:84% they're all called digital assistants. 00:04:17.153 --> 00:04:19.800 align:middle line:84% And they're kind of modeled after the executive assistant. 00:04:19.800 --> 00:04:22.980 align:middle line:84% We talk about Jibo as being a helpful companion. 00:04:22.980 --> 00:04:25.192 align:middle line:84% So that speaks more to the social, emotional, 00:04:25.192 --> 00:04:26.400 align:middle line:90% the relational aspects of it. 00:04:26.400 --> 00:04:27.930 align:middle line:90% But Jibo is also helpful. 00:04:27.930 --> 00:04:29.730 align:middle line:90% He can actually do things. 00:04:29.730 --> 00:04:34.590 align:middle line:84% But it's the way he does things are much more humanistic, I 00:04:34.590 --> 00:04:37.650 align:middle line:84% would say, human-centered in its design, less transactional. 00:04:37.650 --> 00:04:41.220 align:middle line:84% The way he expresses himself through movement 00:04:41.220 --> 00:04:45.730 align:middle line:84% and also just the language he uses, it's much more warm, 00:04:45.730 --> 00:04:46.230 align:middle line:90% I would say. 00:04:46.230 --> 00:04:47.610 align:middle line:90% Like there's cool tech. 00:04:47.610 --> 00:04:49.230 align:middle line:90% Jibo is warm tech. 00:04:49.230 --> 00:04:50.310 align:middle line:90% [MUSIC PLAYING] 00:04:50.310 --> 00:04:52.530 align:middle line:84% SARAH HANSEN: One exciting new application for Jibo 00:04:52.530 --> 00:04:54.840 align:middle line:84% that's currently being explored is 00:04:54.840 --> 00:04:57.900 align:middle line:84% how he might help provide mental health care. 00:04:57.900 --> 00:04:59.400 align:middle line:84% CYNTHIA BREAZEAL: One of the studies 00:04:59.400 --> 00:05:03.420 align:middle line:84% we did, we literally designed an emotional wellness coach 00:05:03.420 --> 00:05:05.160 align:middle line:90% skill for Jibo. 00:05:05.160 --> 00:05:08.010 align:middle line:84% And we deployed Jibo in MIT student dorms. 00:05:08.010 --> 00:05:12.780 align:middle line:84% Because as you probably know, mental health is a huge issue. 00:05:12.780 --> 00:05:15.910 align:middle line:84% I mean, even before the pandemic, mental health, 00:05:15.910 --> 00:05:19.080 align:middle line:84% emotional wellness of students at a lot of top colleges 00:05:19.080 --> 00:05:21.030 align:middle line:90% was really concerning. 00:05:21.030 --> 00:05:25.320 align:middle line:84% So we decided to explore whether Jibo as an emotional wellness 00:05:25.320 --> 00:05:29.280 align:middle line:84% coach could actually help mitigate the stress and anxiety 00:05:29.280 --> 00:05:30.840 align:middle line:84% that students experience when they're 00:05:30.840 --> 00:05:34.470 align:middle line:90% at an elite university like MIT. 00:05:34.470 --> 00:05:38.040 align:middle line:84% So we developed some best-in-breed therapeutic 00:05:38.040 --> 00:05:41.040 align:middle line:84% interactions that typically a human would do, 00:05:41.040 --> 00:05:45.150 align:middle line:84% like a coach would do with their client or whatever. 00:05:45.150 --> 00:05:47.190 align:middle line:90% So typically it's human-human. 00:05:47.190 --> 00:05:50.130 align:middle line:84% We took these proven methods and we just, 00:05:50.130 --> 00:05:53.460 align:middle line:84% now Jibo is just doing it with the person. 00:05:53.460 --> 00:05:54.840 align:middle line:90% And we published that work. 00:05:54.840 --> 00:05:56.257 align:middle line:84% And basically the punch line is we 00:05:56.257 --> 00:05:58.200 align:middle line:84% were able to show that, yeah, Jibo seemed 00:05:58.200 --> 00:06:00.340 align:middle line:90% to make a positive difference. 00:06:00.340 --> 00:06:02.910 align:middle line:84% So in terms of the future of health, 00:06:02.910 --> 00:06:04.680 align:middle line:84% people could go into hospitals potentially 00:06:04.680 --> 00:06:06.150 align:middle line:90% and get world-quality care. 00:06:06.150 --> 00:06:08.230 align:middle line:84% But what happens when they go home? 00:06:08.230 --> 00:06:11.640 align:middle line:84% So again, something like a Jibo, or a social robot, 00:06:11.640 --> 00:06:15.228 align:middle line:84% or whatever, the idea that they can extend the quality of care 00:06:15.228 --> 00:06:17.520 align:middle line:84% and support from the doctors and nurses in the hospital 00:06:17.520 --> 00:06:20.940 align:middle line:84% to the home where patients are living every day, that's 00:06:20.940 --> 00:06:22.368 align:middle line:90% a huge unmet need. 00:06:22.368 --> 00:06:23.910 align:middle line:84% And that's a role where potentially I 00:06:23.910 --> 00:06:27.270 align:middle line:84% think an intelligent kind of, again, collaborative 00:06:27.270 --> 00:06:30.840 align:middle line:84% allied agent with this different kind of interaction paradigm 00:06:30.840 --> 00:06:33.548 align:middle line:90% could make a big difference. 00:06:33.548 --> 00:06:35.340 align:middle line:84% SARAH HANSEN: One of the fascinating design 00:06:35.340 --> 00:06:38.520 align:middle line:84% decisions behind Jibo was choosing just 00:06:38.520 --> 00:06:40.950 align:middle line:90% how human to make him. 00:06:40.950 --> 00:06:43.440 align:middle line:84% CYNTHIA BREAZEAL: When we developed Jibo, 00:06:43.440 --> 00:06:45.900 align:middle line:84% we were very intentional in saying 00:06:45.900 --> 00:06:49.140 align:middle line:90% there's the character of Jibo. 00:06:49.140 --> 00:06:52.380 align:middle line:84% When you talk about the perception of how 00:06:52.380 --> 00:06:55.020 align:middle line:84% children or adults perceive these agents, 00:06:55.020 --> 00:06:59.640 align:middle line:84% Jibo as the character was very aware that Jibo was a robot. 00:06:59.640 --> 00:07:02.582 align:middle line:84% And if you tried to ask Jibo things that really 00:07:02.582 --> 00:07:04.290 align:middle line:84% a robot would have no business answering, 00:07:04.290 --> 00:07:06.790 align:middle line:84% like things around religion or whatnot, Jibo would say this, 00:07:06.790 --> 00:07:08.002 align:middle line:90% like hey, I'm just a robot. 00:07:08.002 --> 00:07:09.460 align:middle line:84% You know, that's probably something 00:07:09.460 --> 00:07:11.502 align:middle line:84% that'd be good to talk about another person with. 00:07:11.502 --> 00:07:12.720 align:middle line:90% But I'm just a robot. 00:07:12.720 --> 00:07:15.870 align:middle line:84% So Jibo was very upfront and transparent 00:07:15.870 --> 00:07:17.040 align:middle line:90% about his robot-ness. 00:07:17.040 --> 00:07:18.900 align:middle line:90% [LAUGHS] 00:07:18.900 --> 00:07:22.200 align:middle line:84% SARAH HANSEN: Collaborative allied agents like Jibo 00:07:22.200 --> 00:07:25.830 align:middle line:84% have huge potential to improve the lives of humans. 00:07:25.830 --> 00:07:29.400 align:middle line:84% But as Cynthia pointed out, they can't replace the power 00:07:29.400 --> 00:07:32.930 align:middle line:90% of real human social networks. 00:07:32.930 --> 00:07:34.790 align:middle line:84% CYNTHIA BREAZEAL: This gets into the ethics 00:07:34.790 --> 00:07:36.582 align:middle line:84% and the responsible design of these things. 00:07:36.582 --> 00:07:40.400 align:middle line:84% I think, first and foremost, people need people. 00:07:40.400 --> 00:07:44.000 align:middle line:84% We need our human social networks, our in-person, 00:07:44.000 --> 00:07:47.270 align:middle line:84% caring human social networks to come around us 00:07:47.270 --> 00:07:49.018 align:middle line:90% when we're in trouble. 00:07:49.018 --> 00:07:50.810 align:middle line:84% Social media companies will try to tell you 00:07:50.810 --> 00:07:52.602 align:middle line:84% your network of 5,000 people you don't know 00:07:52.602 --> 00:07:55.610 align:middle line:84% is just as important as your actual, real, in-person 00:07:55.610 --> 00:07:57.200 align:middle line:90% relationships. 00:07:57.200 --> 00:08:00.140 align:middle line:84% Real, in-person relationships are absolutely essential. 00:08:00.140 --> 00:08:02.165 align:middle line:84% They're essential for our well-being. 00:08:02.165 --> 00:08:04.040 align:middle line:84% They're essential for children's development. 00:08:04.040 --> 00:08:05.290 align:middle line:90% I mean, they're just critical. 00:08:05.290 --> 00:08:08.270 align:middle line:84% And so it signals, I think we're living in this time 00:08:08.270 --> 00:08:12.530 align:middle line:84% when people's time and attention has become 00:08:12.530 --> 00:08:14.540 align:middle line:84% this incredibly precious resource where 00:08:14.540 --> 00:08:19.010 align:middle line:84% the fact that someone will stop and take that time out for you 00:08:19.010 --> 00:08:21.860 align:middle line:84% signals something of you're valued. 00:08:21.860 --> 00:08:24.050 align:middle line:84% You are really valued, and you're meaningful. 00:08:24.050 --> 00:08:25.348 align:middle line:90% [MUSIC PLAYING] 00:08:25.348 --> 00:08:26.710 align:middle line:90% 00:08:26.710 --> 00:08:28.450 align:middle line:84% SARAH HANSEN: As a parent, the importance 00:08:28.450 --> 00:08:30.790 align:middle line:84% of real social networks for children 00:08:30.790 --> 00:08:33.070 align:middle line:90% really feels true to me. 00:08:33.070 --> 00:08:35.890 align:middle line:84% But I also know that my child is growing up in a world 00:08:35.890 --> 00:08:38.409 align:middle line:84% where digital interactions are becoming 00:08:38.409 --> 00:08:41.530 align:middle line:84% more and more common and potentially fraught 00:08:41.530 --> 00:08:43.909 align:middle line:90% with ethical implications. 00:08:43.909 --> 00:08:45.910 align:middle line:84% So I shared with Cynthia a recording 00:08:45.910 --> 00:08:51.180 align:middle line:84% of my child talking with our home executive assistant. 00:08:51.180 --> 00:08:54.120 align:middle line:84% CHILD: Alexa, what's your favorite color? 00:08:54.120 --> 00:08:55.890 align:middle line:84% AMAZON ALEXA: I like ultraviolet. 00:08:55.890 --> 00:08:57.790 align:middle line:90% It glows with everything. 00:08:57.790 --> 00:09:00.180 align:middle line:84% And here's another fun fact about me. 00:09:00.180 --> 00:09:02.310 align:middle line:84% If you'd like, I can learn your voice 00:09:02.310 --> 00:09:05.970 align:middle line:84% to do all sorts of neat things like greeting you by name. 00:09:05.970 --> 00:09:08.040 align:middle line:90% You'll need your own profile. 00:09:08.040 --> 00:09:10.170 align:middle line:90% I can create one for you now. 00:09:10.170 --> 00:09:12.400 align:middle line:90% What's your name? 00:09:12.400 --> 00:09:14.280 align:middle line:90% CHILD: Alexa, off. 00:09:14.280 --> 00:09:16.290 align:middle line:84% AMAZON ALEXA: I'll create a new profile 00:09:16.290 --> 00:09:18.540 align:middle line:84% and send information about Alexa terms 00:09:18.540 --> 00:09:20.940 align:middle line:84% and our privacy notice to your app. 00:09:20.940 --> 00:09:23.820 align:middle line:84% Should I go ahead and create a profile with the name "Off?" 00:09:23.820 --> 00:09:26.122 align:middle line:90% Parent: "Alexa, off." 00:09:26.122 --> 00:09:27.080 align:middle line:90% CYNTHIA BREAZEAL: Yeah. 00:09:27.080 --> 00:09:34.580 align:middle line:84% So as illustrated by that clip, kids are growing up in a time 00:09:34.580 --> 00:09:37.610 align:middle line:84% when they will have always been able to talk and interact 00:09:37.610 --> 00:09:41.450 align:middle line:84% with intelligent machines, and not just in this verbal way 00:09:41.450 --> 00:09:43.980 align:middle line:84% that we heard in the clip, but hidden ways. 00:09:43.980 --> 00:09:44.480 align:middle line:90% Right? 00:09:44.480 --> 00:09:47.390 align:middle line:84% So when you're on social media and it's 00:09:47.390 --> 00:09:51.890 align:middle line:84% trying to learn kind of what your preferences are, 00:09:51.890 --> 00:09:54.470 align:middle line:84% and giving you recommendations that 00:09:54.470 --> 00:09:57.470 align:middle line:84% are not trying to broaden your perspective but 00:09:57.470 --> 00:10:00.110 align:middle line:84% if anything narrow, and narrow, and narrow, and narrow. 00:10:00.110 --> 00:10:03.800 align:middle line:84% Or of course what we're learning is, what gets your attention? 00:10:03.800 --> 00:10:05.630 align:middle line:90% What tends to get shared? 00:10:05.630 --> 00:10:07.480 align:middle line:90% Things that really upset you. 00:10:07.480 --> 00:10:10.055 align:middle line:84% [LAUGHS] Those are the things that really-- you know? 00:10:10.055 --> 00:10:12.560 align:middle line:84% So that's what it tends to feed you 00:10:12.560 --> 00:10:14.900 align:middle line:90% surprisingly quickly, right? 00:10:14.900 --> 00:10:17.120 align:middle line:84% So it's important for people of all ages, 00:10:17.120 --> 00:10:18.620 align:middle line:84% but I think it's important for kids. 00:10:18.620 --> 00:10:21.230 align:middle line:84% Because the average age of when kids 00:10:21.230 --> 00:10:24.260 align:middle line:84% get their first smartphone, it's like 10 years old. 00:10:24.260 --> 00:10:28.490 align:middle line:84% And social interaction through technology 00:10:28.490 --> 00:10:34.680 align:middle line:84% becomes such a key way they interact with each other. 00:10:34.680 --> 00:10:37.190 align:middle line:84% So the digital mediated communication 00:10:37.190 --> 00:10:39.740 align:middle line:84% is, it just becomes such an important way for them 00:10:39.740 --> 00:10:42.560 align:middle line:84% to connect and express themselves. 00:10:42.560 --> 00:10:47.990 align:middle line:84% And just understanding what's under the hood, how it could 00:10:47.990 --> 00:10:54.350 align:middle line:84% be biasing, shaping their attitudes, potentially shaping 00:10:54.350 --> 00:10:55.440 align:middle line:90% behaviors. 00:10:55.440 --> 00:10:57.170 align:middle line:90% This is what's really critical-- 00:10:57.170 --> 00:10:58.130 align:middle line:90% and why. 00:10:58.130 --> 00:11:01.630 align:middle line:90% Who's doing it and why? 00:11:01.630 --> 00:11:03.580 align:middle line:84% SARAH HANSEN: That's where RAISE comes in. 00:11:03.580 --> 00:11:07.480 align:middle line:84% RAISE, which stands for Responsible AI for Social 00:11:07.480 --> 00:11:11.020 align:middle line:84% Empowerment and Education, is an MIT initiative 00:11:11.020 --> 00:11:14.950 align:middle line:84% focused on exactly these kinds of questions. 00:11:14.950 --> 00:11:17.280 align:middle line:84% CYNTHIA BREAZEAL: It's kind of the cross section of AI 00:11:17.280 --> 00:11:19.960 align:middle line:90% and learning, AI and education. 00:11:19.960 --> 00:11:21.630 align:middle line:84% So how can we develop AI systems that 00:11:21.630 --> 00:11:26.250 align:middle line:84% help us learn, as well as how can we understand how 00:11:26.250 --> 00:11:31.020 align:middle line:84% to best empower people to learn about AI so that they 00:11:31.020 --> 00:11:35.430 align:middle line:84% can be informed citizens in how they use these technologies? 00:11:35.430 --> 00:11:37.920 align:middle line:84% And how do we cultivate the future designers 00:11:37.920 --> 00:11:41.460 align:middle line:84% of these technologies that are responsible and ethical 00:11:41.460 --> 00:11:46.110 align:middle line:84% in their designs, but also a far more diverse and inclusive 00:11:46.110 --> 00:11:47.250 align:middle line:90% future workforce? 00:11:47.250 --> 00:11:50.100 align:middle line:84% We love to position the student as the designer 00:11:50.100 --> 00:11:53.160 align:middle line:84% of these technologies, so as the active person who's 00:11:53.160 --> 00:11:56.050 align:middle line:84% thinking about the design, the social implications, the goals, 00:11:56.050 --> 00:11:57.210 align:middle line:90% all of that. 00:11:57.210 --> 00:11:59.640 align:middle line:84% It really empowers students in a way 00:11:59.640 --> 00:12:03.690 align:middle line:84% that they lean in, and discuss and are invested in it. 00:12:03.690 --> 00:12:06.270 align:middle line:84% I think you want children to grow up, 00:12:06.270 --> 00:12:08.250 align:middle line:84% students to grow up with the attitude 00:12:08.250 --> 00:12:10.530 align:middle line:84% that I actually can shape these things. 00:12:10.530 --> 00:12:14.970 align:middle line:84% This is a world that I can be in, and belong in, and shape. 00:12:14.970 --> 00:12:17.100 align:middle line:84% We have both a research kind of objective. 00:12:17.100 --> 00:12:20.940 align:middle line:84% We have an outreach, k-12 outreach, objective. 00:12:20.940 --> 00:12:23.790 align:middle line:84% And a lot of that is bringing these materials 00:12:23.790 --> 00:12:26.430 align:middle line:84% to teachers and students of underrepresented, 00:12:26.430 --> 00:12:28.350 align:middle line:84% under-resourced schools and communities. 00:12:28.350 --> 00:12:31.830 align:middle line:84% Because the field is not diverse or inclusive. 00:12:31.830 --> 00:12:34.350 align:middle line:84% We need to take action to correct that. 00:12:34.350 --> 00:12:37.440 align:middle line:84% And so we have an explicit diversity and inclusion 00:12:37.440 --> 00:12:40.380 align:middle line:84% mission in what we're doing through the outreach. 00:12:40.380 --> 00:12:45.120 align:middle line:84% And we are trying to design the curriculum, the concepts, 00:12:45.120 --> 00:12:47.850 align:middle line:84% the activities to meet students where 00:12:47.850 --> 00:12:50.970 align:middle line:84% they are in things that they use and they think about so it's 00:12:50.970 --> 00:12:53.340 align:middle line:90% super relevant to them now. 00:12:53.340 --> 00:12:57.450 align:middle line:84% But then we also expose them to the way AI is transforming 00:12:57.450 --> 00:13:00.720 align:middle line:84% so many disciplines, and markets, and industries 00:13:00.720 --> 00:13:02.700 align:middle line:84% so they understand it's here and it's 00:13:02.700 --> 00:13:05.863 align:middle line:84% going to continue to shape our futures. 00:13:05.863 --> 00:13:07.530 align:middle line:84% SARAH HANSEN: Part of that work involves 00:13:07.530 --> 00:13:11.040 align:middle line:84% offering an opportunity for young people and their teachers 00:13:11.040 --> 00:13:13.710 align:middle line:84% across the country to participate in the upcoming 00:13:13.710 --> 00:13:18.237 align:middle line:90% Day of AI on May 13th, 2022. 00:13:18.237 --> 00:13:19.820 align:middle line:84% CYNTHIA BREAZEAL: We develop all kinds 00:13:19.820 --> 00:13:24.380 align:middle line:84% of curriculum of different formats and different lengths. 00:13:24.380 --> 00:13:28.580 align:middle line:84% So Day of AI is the introductory experience. 00:13:28.580 --> 00:13:32.720 align:middle line:84% And it's really an opportunity, if teachers or students 00:13:32.720 --> 00:13:37.040 align:middle line:84% are just curious about AI, it is literally a program 00:13:37.040 --> 00:13:40.610 align:middle line:84% intended to be for any kind of teacher, 00:13:40.610 --> 00:13:43.910 align:middle line:84% for students from elementary up to high school, 00:13:43.910 --> 00:13:49.100 align:middle line:84% where they can learn about AI from more 00:13:49.100 --> 00:13:52.130 align:middle line:84% of the lens of its implications on digital literacy 00:13:52.130 --> 00:13:55.770 align:middle line:84% and digital citizenship, but also demystifying it as well. 00:13:55.770 --> 00:13:58.580 align:middle line:84% So all great bands kind of have this, 00:13:58.580 --> 00:14:00.870 align:middle line:90% what is AI introductory module? 00:14:00.870 --> 00:14:03.590 align:middle line:84% So for instance, like when I spoke about this AI 00:14:03.590 --> 00:14:07.040 align:middle line:84% and creativity curriculum, we've had students who are saying, 00:14:07.040 --> 00:14:09.890 align:middle line:84% I didn't think I was relevant or important to me 00:14:09.890 --> 00:14:12.470 align:middle line:84% because I want to be a photographer when I grow up. 00:14:12.470 --> 00:14:15.020 align:middle line:84% But they're like, now I realize, wow, 00:14:15.020 --> 00:14:16.970 align:middle line:84% AI could actually be a really cool tool that I 00:14:16.970 --> 00:14:19.765 align:middle line:84% can use creatively in my profession of photography 00:14:19.765 --> 00:14:20.390 align:middle line:90% when I grow up. 00:14:20.390 --> 00:14:24.290 align:middle line:84% So a lot of this just to help kids to understand that chances 00:14:24.290 --> 00:14:27.020 align:middle line:84% are, this is something that's going to be 00:14:27.020 --> 00:14:29.130 align:middle line:90% empowering and relevant to you. 00:14:29.130 --> 00:14:32.450 align:middle line:84% So just be aware of it, being able to leverage 00:14:32.450 --> 00:14:34.760 align:middle line:84% it to help you achieve your goals 00:14:34.760 --> 00:14:36.338 align:middle line:90% and your ambitions in life. 00:14:36.338 --> 00:14:37.130 align:middle line:90% SARAH HANSEN: Well. 00:14:37.130 --> 00:14:39.200 align:middle line:84% If there's a teacher out there like me 00:14:39.200 --> 00:14:42.650 align:middle line:84% who sometimes can't even get her agent to turn the lights on, 00:14:42.650 --> 00:14:44.202 align:middle line:90% is this still appropriate? 00:14:44.202 --> 00:14:45.410 align:middle line:90% CYNTHIA BREAZEAL: Absolutely. 00:14:45.410 --> 00:14:47.400 align:middle line:84% So it's for all kinds of teachers. 00:14:47.400 --> 00:14:47.900 align:middle line:90% Right? 00:14:47.900 --> 00:14:51.260 align:middle line:84% This is not a technical, computer science, 00:14:51.260 --> 00:14:53.000 align:middle line:90% STEM curriculum at all. 00:14:53.000 --> 00:14:55.010 align:middle line:84% It's really much more around, again, 00:14:55.010 --> 00:14:59.090 align:middle line:84% digital citizenship and digital literacy as it pertains to AI, 00:14:59.090 --> 00:15:02.180 align:middle line:84% so again, AI in our lives, and understanding that, 00:15:02.180 --> 00:15:05.030 align:middle line:84% and having conversations about that, 00:15:05.030 --> 00:15:07.260 align:middle line:84% and designing some cool things around that. 00:15:07.260 --> 00:15:07.760 align:middle line:90% [LAUGHS] 00:15:07.760 --> 00:15:08.660 align:middle line:90% SARAH HANSEN: Yeah. 00:15:08.660 --> 00:15:11.660 align:middle line:84% So where can teachers go to sign up? 00:15:11.660 --> 00:15:13.968 align:middle line:90% And is there a fee associated? 00:15:13.968 --> 00:15:15.260 align:middle line:90% CYNTHIA BREAZEAL: So it's free. 00:15:15.260 --> 00:15:17.150 align:middle line:84% Again, one of the great things about MIT 00:15:17.150 --> 00:15:20.940 align:middle line:84% is we want to lower the barrier as much as humanly possible. 00:15:20.940 --> 00:15:22.620 align:middle line:90% So all the materials are free. 00:15:22.620 --> 00:15:25.280 align:middle line:90% You can go to DayofAI.org. 00:15:25.280 --> 00:15:28.500 align:middle line:84% You can look at the curriculum right there. 00:15:28.500 --> 00:15:31.100 align:middle line:84% We also offer free teacher training. 00:15:31.100 --> 00:15:32.990 align:middle line:84% You can register and you can sign up 00:15:32.990 --> 00:15:34.940 align:middle line:90% for an online training session. 00:15:34.940 --> 00:15:37.880 align:middle line:84% We are collaborating with I2 Learning. 00:15:37.880 --> 00:15:40.280 align:middle line:84% This is a company that what they do is train teachers. 00:15:40.280 --> 00:15:41.280 align:middle line:90% This is what they do. 00:15:41.280 --> 00:15:45.860 align:middle line:84% And so we've worked with them to take our curriculum materials, 00:15:45.860 --> 00:15:48.980 align:middle line:84% and using kind of their experience to really tune it 00:15:48.980 --> 00:15:51.030 align:middle line:84% for what they think is great for, again, 00:15:51.030 --> 00:15:53.300 align:middle line:84% a broad, general teacher audience. 00:15:53.300 --> 00:15:56.330 align:middle line:84% And they are running the teacher training sessions. 00:15:56.330 --> 00:15:58.470 align:middle line:90% And again, it's completely free. 00:15:58.470 --> 00:16:01.830 align:middle line:84% So we would love for teachers to sign up. 00:16:01.830 --> 00:16:04.400 align:middle line:84% We are in almost all 50 states now. 00:16:04.400 --> 00:16:05.630 align:middle line:90% [LAUGHS] 00:16:05.630 --> 00:16:07.460 align:middle line:84% SARAH HANSEN: RAISE and the Day of AI 00:16:07.460 --> 00:16:10.790 align:middle line:84% are increasing access to AI education. 00:16:10.790 --> 00:16:14.150 align:middle line:84% And they're also informing how we teach it. 00:16:14.150 --> 00:16:16.190 align:middle line:84% CYNTHIA BREAZEAL: It is an open question 00:16:16.190 --> 00:16:21.080 align:middle line:84% still about what are the most effective ways of educating 00:16:21.080 --> 00:16:24.080 align:middle line:84% students at a variety of ages and backgrounds 00:16:24.080 --> 00:16:26.360 align:middle line:84% about what artificial intelligence is 00:16:26.360 --> 00:16:27.800 align:middle line:90% and how it works. 00:16:27.800 --> 00:16:29.510 align:middle line:84% We still have much we need to understand 00:16:29.510 --> 00:16:32.000 align:middle line:84% in terms of what grades, what preparation, 00:16:32.000 --> 00:16:35.772 align:middle line:84% what concepts, how can we demystify 00:16:35.772 --> 00:16:37.730 align:middle line:84% and these computational methods in a way that's 00:16:37.730 --> 00:16:39.380 align:middle line:90% grade and age-appropriate? 00:16:39.380 --> 00:16:41.240 align:middle line:84% How we can prepare teachers-- there's 00:16:41.240 --> 00:16:42.990 align:middle line:84% a whole bunch of research questions there. 00:16:42.990 --> 00:16:44.810 align:middle line:84% What are the right approaches, methods 00:16:44.810 --> 00:16:47.455 align:middle line:84% of professional development for teachers who, I mean, 00:16:47.455 --> 00:16:48.080 align:middle line:90% I can tell you. 00:16:48.080 --> 00:16:51.980 align:middle line:84% They didn't major in AI in college, right? 00:16:51.980 --> 00:16:54.080 align:middle line:90% So this is new to them too. 00:16:54.080 --> 00:16:56.360 align:middle line:84% So how do we do that effectively? 00:16:56.360 --> 00:16:58.610 align:middle line:84% What are the longer impacts of this 00:16:58.610 --> 00:17:01.700 align:middle line:84% in terms of attitudes and dispositions 00:17:01.700 --> 00:17:04.010 align:middle line:84% that students may form about these technologies, 00:17:04.010 --> 00:17:06.710 align:middle line:84% and about themselves, and their relationship 00:17:06.710 --> 00:17:08.569 align:middle line:90% to these technologies? 00:17:08.569 --> 00:17:12.470 align:middle line:84% And then as we do this, how do we again holistically develop 00:17:12.470 --> 00:17:13.710 align:middle line:90% the whole person as well? 00:17:13.710 --> 00:17:16.349 align:middle line:84% So teamwork, collaboration, communication, 00:17:16.349 --> 00:17:19.190 align:middle line:90% critical thinking, creative-- 00:17:19.190 --> 00:17:22.760 align:middle line:84% they can all be kind of woven into this tapestry of how 00:17:22.760 --> 00:17:27.470 align:middle line:84% we try to address AI literacy through positioning 00:17:27.470 --> 00:17:29.450 align:middle line:84% students as the designers, and creators, 00:17:29.450 --> 00:17:32.580 align:middle line:84% and makers of these solutions with these technologies. 00:17:32.580 --> 00:17:36.230 align:middle line:90% So we design these activities. 00:17:36.230 --> 00:17:38.010 align:middle line:90% And then we evaluate them. 00:17:38.010 --> 00:17:38.510 align:middle line:90% Right? 00:17:38.510 --> 00:17:42.810 align:middle line:84% So we will reach out and recruit students and teachers. 00:17:42.810 --> 00:17:44.300 align:middle line:90% And we will engage them. 00:17:44.300 --> 00:17:46.440 align:middle line:84% We often engage in co-design with them. 00:17:46.440 --> 00:17:48.410 align:middle line:84% So even as we're developing these materials, 00:17:48.410 --> 00:17:50.390 align:middle line:84% we will work with teachers and students 00:17:50.390 --> 00:17:54.950 align:middle line:84% as we develop them to make sure that they're effective, 00:17:54.950 --> 00:17:56.690 align:middle line:90% and responding, and engaging. 00:17:56.690 --> 00:17:59.840 align:middle line:84% And then we do rigorous evaluations 00:17:59.840 --> 00:18:02.990 align:middle line:84% of what are they understanding or not understanding? 00:18:02.990 --> 00:18:04.922 align:middle line:84% What is their pathway to understanding? 00:18:04.922 --> 00:18:06.380 align:middle line:84% And what is the implication of that 00:18:06.380 --> 00:18:09.380 align:middle line:84% for how we think about staging these concepts and ideas? 00:18:09.380 --> 00:18:13.557 align:middle line:84% Are we seeing evidence that they are developing dispositions 00:18:13.557 --> 00:18:15.140 align:middle line:84% about AI that are empowering for them, 00:18:15.140 --> 00:18:17.800 align:middle line:84% or do they find it as being intimidating or not for them? 00:18:17.800 --> 00:18:21.100 align:middle line:84% So there's a lot of questions that we're trying to assess. 00:18:21.100 --> 00:18:24.430 align:middle line:84% And then when we have something we think is solid, 00:18:24.430 --> 00:18:26.990 align:middle line:84% then we make it available for free. 00:18:26.990 --> 00:18:28.360 align:middle line:90% It's all Creative Commons. 00:18:28.360 --> 00:18:34.030 align:middle line:84% Our goal is just to help empower teachers, non-profits, 00:18:34.030 --> 00:18:37.210 align:middle line:84% get these materials out there and share them freely. 00:18:37.210 --> 00:18:38.890 align:middle line:84% SARAH HANSEN: But again, no matter how 00:18:38.890 --> 00:18:42.040 align:middle line:84% good robots or AI get, they're no replacement 00:18:42.040 --> 00:18:44.020 align:middle line:90% for the real thing. 00:18:44.020 --> 00:18:45.850 align:middle line:84% CYNTHIA BREAZEAL: I think a lot of AI 00:18:45.850 --> 00:18:47.762 align:middle line:84% has assumed when you're trying to build 00:18:47.762 --> 00:18:49.970 align:middle line:84% the social and emotional qualities into technologies, 00:18:49.970 --> 00:18:51.910 align:middle line:90% it's about the robot. 00:18:51.910 --> 00:18:54.310 align:middle line:84% And I'm like, no, it's about the people 00:18:54.310 --> 00:18:57.670 align:middle line:84% because this is what people need to engage. 00:18:57.670 --> 00:19:01.510 align:middle line:84% The more deeply we can engage of our holistic self, 00:19:01.510 --> 00:19:04.203 align:middle line:84% not just our cognitive selves, but our social selves, 00:19:04.203 --> 00:19:06.370 align:middle line:84% our emotional selves, our physical selves-- the more 00:19:06.370 --> 00:19:09.100 align:middle line:84% deeply we can engage, not surprisingly, 00:19:09.100 --> 00:19:11.720 align:middle line:84% the more successful we are, the better our outcomes. 00:19:11.720 --> 00:19:13.810 align:middle line:84% So it's to say, it's not about let's 00:19:13.810 --> 00:19:16.810 align:middle line:84% build machines that look like people and act like people. 00:19:16.810 --> 00:19:20.110 align:middle line:84% It's, let's build machines that dovetail with us in a way 00:19:20.110 --> 00:19:22.900 align:middle line:84% that we can bring forth all of ourselves 00:19:22.900 --> 00:19:24.670 align:middle line:84% to help us achieve the goals that 00:19:24.670 --> 00:19:26.360 align:middle line:90% are really important to us. 00:19:26.360 --> 00:19:30.518 align:middle line:84% And this is why when we talk about AI as pedagogical agents, 00:19:30.518 --> 00:19:32.560 align:middle line:84% I'm like, we will never get rid of human teachers 00:19:32.560 --> 00:19:35.920 align:middle line:84% because they are essential for children 00:19:35.920 --> 00:19:40.330 align:middle line:84% to know that in this community of human beings, 00:19:40.330 --> 00:19:45.250 align:middle line:84% adults invest their time in children's learning 00:19:45.250 --> 00:19:46.090 align:middle line:90% and development. 00:19:46.090 --> 00:19:47.410 align:middle line:90% Like, you matter. 00:19:47.410 --> 00:19:49.150 align:middle line:84% I'm taking time and energy to do this 00:19:49.150 --> 00:19:50.317 align:middle line:90% because it's that important. 00:19:50.317 --> 00:19:51.500 align:middle line:90% And you are that important. 00:19:51.500 --> 00:19:54.054 align:middle line:84% You know, you just can't underestimate that. 00:19:54.054 --> 00:19:56.710 align:middle line:90% [MUSIC PLAYING] 00:19:56.710 --> 00:19:59.830 align:middle line:84% SARAH HANSEN: To take part in the Day of AI on May 13, 00:19:59.830 --> 00:20:02.380 align:middle line:90% go to DayofAI.org. 00:20:02.380 --> 00:20:05.440 align:middle line:84% You can access free teacher training and other Creative 00:20:05.440 --> 00:20:08.380 align:middle line:84% Commons educational materials at RAISE.MIT.edu. 00:20:08.380 --> 00:20:14.128 align:middle line:84% Or you can access them through our website at OCW.MIT.edu. 00:20:14.128 --> 00:20:16.930 align:middle line:90% 00:20:16.930 --> 00:20:19.690 align:middle line:84% You can also find Cynthia's free and open teaching materials 00:20:19.690 --> 00:20:22.820 align:middle line:84% from her MIT courses on our website. 00:20:22.820 --> 00:20:25.010 align:middle line:90% Thank you so much for listening. 00:20:25.010 --> 00:20:29.380 align:middle line:84% Until next time, signing off from a galaxy not too far away, 00:20:29.380 --> 00:20:33.490 align:middle line:84% I'm Sarah Hansen from MIT OpenCourseWare. 00:20:33.490 --> 00:20:37.420 align:middle line:84% Chalk Radio's producers include myself, Brett Paci, 00:20:37.420 --> 00:20:39.100 align:middle line:90% and Dave Lishansky. 00:20:39.100 --> 00:20:42.400 align:middle line:84% Scriptwriting assistance from Aubrey Calaway. 00:20:42.400 --> 00:20:46.150 align:middle line:84% Show notes for this episode were written by Peter Chipman. 00:20:46.150 --> 00:20:50.350 align:middle line:84% The RAISE online publication was shared on MIT OpenCourseWare 00:20:50.350 --> 00:20:51.940 align:middle line:90% by Sharon Lin. 00:20:51.940 --> 00:20:55.480 align:middle line:84% We're funded by MIT Open Learning and supporters 00:20:55.480 --> 00:20:56.440 align:middle line:90% like you. 00:20:56.440 --> 00:20:58.890 align:middle line:90% [MUSIC PLAYING] 00:20:58.890 --> 00:21:03.096 align:middle line:90%