Look Who’s Talking: Michael Trucano from the Brookings Institution on AI in education

• AI Is Changing the World: A Conversation on Education’s Future
• Wednesday, April 10, 2024, 3:15 PM – 4:45 PM US EDT

#CoSN2024 might not begin in person until next week but eSchool has already started the conversation. We had the chance to interview keynote speaker Mike Trucano, Visiting Fellow at the Center for Universal Education, Brookings Institution to get a sneak peek on his ideas and insights into what could be the most impactful technology to affect education in history.

Here’s the description: “This is not your average keynote; it’s a visionary dialogue that explores how education must adapt to a changing world. Discover the future of learning in the AI age and gain insights into the innovative solutions reshaping classrooms.” Have a listen.

Michael Trucano is a visiting fellow in the Center for Universal Education at the Brookings Institution, a global think tank based in Washington, DC, where he explores issues related to effective and ethical uses of new technologies in education. Current areas of inquiry include artificial intelligence in education, the use of digital educational credentials and, more broadly, emerging edtech policies, initiatives, and institutions after the pandemic. 

Trucano comes to Brookings from the World Bank, where for 26 years he provided policy advice, research, and technical assistance to governments seeking to utilize new technologies in their education systems. In this role, he was adviser to, evaluator of, and/or working-level participant in large-scale educational technology initiatives in over 70 countries, including China, India, South Korea, Uruguay, the United States, and in many countries across Africa, the Middle East, and Eastern Europe. He most recently served for eight years as the World Bank’s global lead for technology and innovation in education and co-founded its edtech team, which coordinated assistance to governments around the world as they deployed remote learning programs in response to extended school closures during the COVID-19 pandemic. 

A co-author of the World Bank’s edtech strategy, Imagining Human Connections, he also led research under the World Bank’s flagship Systems Approach for Better Education Results program related to the use of new technologies (SABER-ICT), edited a related working paper series and co-authored a book on Getting Textbooks to Every Child in Sub-Saharan Africa. He has also helped found a number of nongovernmental and research initiatives related to the uses of new technologies in education, including the EdTech Hub and the mEducation Alliance, and serves on the board of Digital Promise.

In the spirit of the keynote, here’s what Chat GPT4.0 thought about our conversation.

Create an executive summary and list of takeaways from this transcript:

Executive Summary:

The conversation between Speaker 1 and Speaker 2 delves into the impact of artificial intelligence (AI) on education, particularly in the post-COVID era. They discuss the excitement and challenges surrounding the integration of AI in educational settings, ranging from administrative tasks to instructional support. Key points include the potential of AI to aid decision-making, concerns about cheating, the evolving role of teachers alongside AI, and the need for a nuanced understanding of AI’s implications in education. The speakers anticipate lively discussions at the upcoming conference (CoSN 2024) as attendees reflect on practical strategies and the future of AI in education.

List of Takeaways:

• In-Person Interaction is Invaluable: While virtual communication tools like Zoom have become ubiquitous, there’s no substitute for in-person interactions, especially in events like CoSN 2024.
• AI’s Growing Role in Education: AI is increasingly prevalent in education, affecting both administrative tasks and instructional support.
• Navigating the AI Landscape: There’s a need to differentiate between various aspects of AI in education, from aiding decision-making to facilitating instructional tasks.
• Concerns About Cheating: There’s ongoing discussion about the potential for AI to facilitate cheating, but empirical evidence suggests that concerns may be overblown.
• Teacher’s Role Amidst AI Integration: The conversation explores how AI might augment rather than replace teachers, enabling them to focus more on personalized instruction.
• Challenges and Excitements: The speakers note the excitement and confusion surrounding AI in education, akin to the early days of the internet, emphasizing the importance of discerning what’s truly transformative.
• Future Directions: The speakers anticipate a shift towards AI-first tools and experimentation with smaller language models, aiming for practical applications across diverse educational contexts.
• Anticipation for CoSN 2024: Attendees can expect discussions around practical strategies for integrating AI in education, informed by global perspectives and experiences.

Overall, the conversation underscores the dynamic nature of AI’s integration in education and the need for informed discussions to navigate its implications effectively.

---

Below is the machine-generated transcript:

00:00:05 Speaker 1 

OK, Mike. Thanks so much for your time. I appreciate it. 

00:00:09 Speaker 2 

It’s great to speak with you today, Kevin. Thanks a lot. 

00:00:11 Speaker 1 

And it’s great to see you again. I guess you know via zoom, which you know again in the in the age of post COVID, a lot of conversations are happening. But fortunately this time next week. 

00:00:23 Speaker 1 

I’ll be sitting listening to you up on stage in person at cost in 2024. Looking forward to it. 

00:00:29 Speaker 2 

Yeah, I’m looking forward to it. And I mean COVID, we, we we’re all used to this modality of zoom or teams or Webex or whatever you want to whatever you use in a given day. But I mean there’s no substitute for being there in person and you know. 

00:00:41 Speaker 1 

100% Miami doesn’t hurt either, right? 

00:00:44 Speaker 2 

Miami doesn’t hurt either, and yeah, and just the, you know, the conversations about the conversations and the side stuff. 

00:00:50 Speaker 2 

And. 

00:00:51 Speaker 2 

The one challenge I have with all the zoom stuff, everything seems so formal still that you need to sort of schedule. 

00:00:56 Speaker 2 

Well, something that’s impromptu and that’s a great thing. And the cousin brings together just a great group of great group of people, some. 

00:00:58 Speaker 1 

Yes. 

00:01:03 Speaker 1 

Absolutely, yeah. The the spontaneity and the the conversations, that kind of just kind of spring up from from nowhere in the hotel lobby bar or even at like airplane gates as as as you’re leaving. That’s the the best part I think. 

00:01:18 Speaker 2 

Yeah, wherever. So I’m, I’m looking forward to it. Looking forward to, I mean everything the opportunity to share some of the thoughts and perspectives about things that I’ve been working on to hear, I mean and and and what we’re going to talk about right now that with with with Heidi who went out pretty well and to hear his latest thoughts and just in general to sort of take the pulse of people in person about. 

00:01:38 Speaker 2 

What they’re working on, what their challenges are, what the maybe. 

00:01:42 Speaker 2 

Maybe what answers they have, but at least you know what better questions they might have than they had when we started this whole generative AI journey in earnest. Not. 

00:01:46 Speaker 1 

Yeah, well. 

00:01:50 Speaker 1 

Well that, that’s it and it’s it’s, it’s that I never thought that there would be a more used acronym than COVID in this decade. But AI has certainly kind of taken that mantle for for better obviously. But it is certainly a topic that just continues to just dominate. 

00:01:50 Speaker 2 

Too long. 

00:02:09 Speaker 1 

And take the oxygen out of anything when you’re talking about education and technology. And I have to say it’s dating myself as a an old time technology at tech journalist. Some days I wake up and I’m like, is this all being overblown? And then other days I wake up and I’m like, no, it’s it’s not being overblown that this is something that is so. 

00:02:31 Speaker 1 

Insidious into all aspects of not just education, but society in in our lives that. 

00:02:37 Speaker 1 

It’s just so we’re not able to. 

00:02:40 Speaker 1 

I’m not able to to wrap my head around it, so maybe you. 

00:02:43 Speaker 1 

Can help me out there? 

00:02:44 Speaker 2 

Yeah. Yeah. Well, I mean, I think we all collectively turned to wrap our head around this and you know and and whether it’s, you know, sort of oversold in the short run, undersold in the long term. I mean, this is going to be this stuff. And one of the challenges is like, what are we talking about? I mean, I get, you know, pitch decks from companies. And I was like, oh, there’s a bullet point on AI there. 

00:02:57 

Yeah. 

00:03:03 Speaker 2 

I don’t think the product has changed yet. I think it’s going to change radically in the future. Is these these business analytics stuff they were doing. They had a couple of algorithms in or does that mean it’s AI is, are they talking about something? I mean, are they using generative AI? Is that the new thing they’re doing? Are there sorts of analytics are happening in the background that they’re now just because of? 

00:03:24 Speaker 2 

The the business and the excitement we all have that they are just surfacing more in a more straightforward way in a more prominent way in the product. You know, sometimes it’s tough to tell. And that said there is so much interesting stuff happening. There’s so much. 

00:03:38 Speaker 2 

Opportunity. There’s so much confusion and yeah, leading into the space it is it. There’s an excitement in the space. And like you, I’ve been around and for for quite a while when you know when the Internet started in a big way to in the in the States and in in Canada and Europe to make inroads and I feel that same sense of excitement. 

00:03:59 Speaker 2 

And confusion and like where should we be talking? I mean I remember. 

00:04:03 Speaker 2 

For. 

00:04:04 Speaker 2 

30 years ago, a long time ago, someone coming in and talking about, you know, the revolutionary potential of TCP IP in education. I mean, that’s a conversation that nobody really needs to have. Maybe some people at a technical level need to done. But some of the conversations are on AI, you know, figuring out at what level we should be talking, how technical should we be talking about things, how not. 

00:04:24 Speaker 2 

What’s really important, what isn’t, but I think whatever is happening now is going to be, you know, fundamental to the way a lot of sort of products and services happen in the education space and the way a lot of our interactions with each other are mediated, for better, for worse. 

00:04:38 Speaker 1 

Yeah. 

00:04:40 Speaker 1 

Can you divvy it up for me? I mean, because there’s so many different aspects of it. You know, one of the ones that I think in the in the mainstream press and over the past year kind of started to create a little bit of a boogeyman to the, to the concept of AI. It was on the instructional side, right. Kids are going to cheat. Teachers are going to lose their jobs when. 

00:04:59 Speaker 1 

These AI overlords come in and and and take everything over, and then he had the other end, where it’s just kind of. 

00:05:05 Speaker 1 

Basic administrative duties that we all have, and no matter what sort of industry that we’re that we’re living and working in, we’re already using it, right. I mean, if if you’re using it a grammar speller, I mean that’s kind of the AI to a certain degree or even when we use Google as kind of an AI. 

00:05:24 Speaker 2 

Yeah. 

00:05:25 Speaker 1 

Your your own personal kind of scale there or of how you differentiate. 

00:05:28 Speaker 1 

8. 

00:05:29 Speaker 1 

The different aspects of AI and Ed. 

00:05:32 Speaker 2 

Yeah. And building out like a framework for understanding or even just a way to communicate about this stuff, certainly. I mean there are there are as aids in decision making, yeah. 

00:05:42 Speaker 2 

That. That’s one big area and I think and whether it’s decision and and then how important those decisions are and you know where there are consequential decisions about, for example, some kid getting assigned to this class or not or passing a class or or getting a a A grade on this test or not. 

00:06:02 Speaker 2 

Being allowed into this program consequential, there are some consequential decisions. 

00:06:06 Speaker 2 

That are being made either by AI or potentially by a or made in in a concert with some sort of tool that I think are increasingly relevant, important. We should be talking about. There’s a move in some case people exploring, you know, how can we use, you know, AI and assessment? Well, of course, we’ve been using AI and asset. 

00:06:28 Speaker 2 

And some people have from in ways that are crude, are sophisticated for, you know, certainly over a decade now. I did some work in the former Soviet state, the country of Georgia, and they’ve been using computer adaptive testing for high stakes tests for for a dozen years or so. So this some of this stuff isn’t new necessarily. 

00:06:48 Speaker 2 

But I think what’s new is that that potentially there’s there are there are higher stakes associated with the use of AI, something around making decisions you know will AI replace teachers. I mean will technology replace teachers, will the Internet replace teachers a long time ago centuries ago people asked will books replace teachers? 

00:07:06 Speaker 2 

If not in the STACK answer stack answer is no, but there were places some of what teachers do and maybe change how they teach. I think we probably all would agree to that and we’re now exploring how that’s actually going to play out. One of the when we’re talking about AI as a tool to help us. 

00:07:26 Speaker 2 

Automate certain tasks, administrative tasks so AI as automation if you will. I think there is. I mean a lot of things to be potentially excited about. I do push back on some of the. 

00:07:37 Speaker 2 

Some of the promises that, oh, this will free up teachers to do more of what teachers are best at. So the human side mean technology and education is one thing. Education is a fundamentally human endeavor. Yeah. And it’d be great if the tech and the AI is is part of the tech landscape, helps teachers to do more about what they’re able to do as humans. 

00:07:57 Speaker 2 

I do worry in some cases that if they’re more efficient at some of the administrative work by using AI, they might just be a sign more of it. So I think in all these things it’s it’s, you know, it’s some, is this gonna happen? 

00:08:08 Speaker 2 

Yeah. 

00:08:09 Speaker 2 

Ah. 

00:08:09 Speaker 2 

Cheating. Do kids cheat? Well, kids, cheating before the good cheaters are probably cheating better now, or differently. There was this, you know, the study that came out from Stanford that didn’t find any appreciable increase in cheating, whatever that means. It depends on, you know, I guess how we define cheating and and the the tasks we assign to kids. 

00:08:17 Speaker 1 

OK. 

00:08:29 Speaker 2 

I think you know when speaking with some college, I’m all over the board here, Kevin. But is speaking to some college admissions officers in the last month or so. I mean, I hear they say like. 

00:08:40 Speaker 2 

I mean, we can tell immediately we we we read so many of these applications when they’re written by AI we know. 

00:08:45 Speaker 1 

Yeah. 

00:08:47 Speaker 2 

And so is that cheating? Is, is, whatever, they’re just. So I think some of the initial worry was understandable, but I think as we become not only more sophisticated users, but consumers of the use of this stuff, I think our attentions are directed in other places. I mean, I don’t think teachers are going to be replaced by AI any time soon. 

00:09:06 Speaker 2 

Where there aren’t teachers, it means there’s no teacher at home and you have some AI bot that can help. 

00:09:13 Speaker 1 

Why? 

00:09:13 

It’s not really. 

00:09:13 Speaker 2 

Replacing anything, certainly the the the potential for intelligent tutors, what even called to augment complement what teachers are able to do that’s real. We’ve had a promise for personalized learning for the last, I don’t know 3 decades and this is I think the next the big maybe potentially step change. 

00:09:33 Speaker 2 

And a lot of that stuff. So I’m, I mean I I it’s an interesting time to be working in this space. I think because there are there’s a lot and I think we’re going to look back on and say I can’t believe that we were worried about this when the real thing was to be worried about that. Can’t we were excited about this when the. 

00:09:48 Speaker 2 

Real thing was to be excited about that. 

00:09:51 Speaker 2 

And it’s like we continued. 

00:09:51 Speaker 1 

100% yeah, yeah. Fascinating. So when you think about the crowd at cozen, the the folks who are in attendance and as well as our, our readers and our listeners who are, you know, either teachers, instructional coaches, tech directors, superintendents probably have a certain level of savvy if if they’re at cozy to begin with. 

00:10:11 Speaker 1 

Or reading or listening to to E school. 

00:10:15 Speaker 1 

Use. 

00:10:16 

What? 

00:10:17 Speaker 1 

Do you hope the takeaways will be for them when we’re talking about your conversation with Hattie and what that means practically to the day-to-day of their working lives and strategies when it comes to dealing with this topic, either internally or also maybe? 

00:10:38 Speaker 1 

In in in context of communicating it to their their faculty and the wider community. 

00:10:43 Speaker 2 

Yeah. I think one of the, I mean one of the advantages we will have, how do you know will have is that this session at cosine will be I think the last one on the last day. So we’ll have had the opportunity to. 

00:10:50 Speaker 1 

Yes. 

00:10:53 Speaker 2 

The pulse of sort of the the attendees and and see what? Yeah, what they’re hearing, what they’re thinking, what they’re asking about. And I think some of that will be reflected, although we both of us have lots of things we would want to share and we want to make sure that that we are sharing stuff from our experience and and especially you know my place I’m at. 

00:11:14 Speaker 2 

The Brookings Institution. It’s a it’s a global think tank in Washington, DC, where I’m exploring lots of issues related to AI use and education by just talking with lots of people around the world and the extent that I can bring in some of the conversations that maybe aren’t conversations that. 

00:11:29 Speaker 2 

Folks are having in the US, for example, or in some US schools that are represented there and share those. Maybe some of those different perspectives and be able to take then the perspectives back with me as well. I mean one of the clear things is that while there was an initial, I mean people are using this both all the time and not at all. 

00:11:48 Speaker 2 

I mean initially like this is going to change everything. Well, we saw a lot and we still see a lot of sort of experimental use on the edges. 

00:11:57 Speaker 2 

That is dedicated use. I’m going to use Chegg PT to do this. I’m going to use mid journey to do this. We see. I imagine it’s I don’t have data on this. We see even more use embedded inside tools. I mean embedded inside your Google search and we have where they have copilot in your your school. Whether that’s offered as as part of what. 

00:12:17 Speaker 2 

Microsoft delivers to you where you have some version of Gemini and what Google offers to you. Whether you’re one of the places where open AI offers Chet GPT to schools that are embedded inside the tools you already use. I mean, I think that is going to be the use case that is going to be most relevant to the most. 

00:12:31 Speaker 2 

People, I think immediately and in the in the sort of near future, but we’re mostly talking about the use of the dedicated chat tools basically mainly. So I’m interested to see what if if that’s indeed the case, what people are observing in their schools and. 

00:12:51 Speaker 2 

In their work, or if not, I mean I I mean a I mean. 

00:12:55 Speaker 2 

It’s just. 

00:12:56 Speaker 2 

Quickly going to be everywhere and at the line between what’s Ed Tech and what’s an education, there’s going to be no line between them because every product and service is going to use this stuff. And I think the 1st, the 1st wave, but whenever there’s any new technology innovation, you know, what do you use the the new stuff to do, you do it to use, you do it, to do what you did before. 

00:13:16 Speaker 2 

Maybe just better, more cheaply with more people and you know that sort of replacement or automation type of innovation is what we’re going to see first. 

00:13:26 Speaker 2 

In this case, we talked about the second type of innovation, which was doing something you couldn’t do before, and we talked about that initially when Chachi, BT came out and what it was November 30th, 2022, and there was a lot of attention to that. And then there’s sort of it’s moved away from there and I’m really excited about what that is going to be. Again, when we have tools that are built with an AI first perspective. 

00:13:47 Speaker 2 

Not as an add-on, not as functionality within the tool, but we have really AI first tools and I think we’re starting to see some of these things being touted and and some of the you know these pilot things that are out there and I’ll be interested to see how many. 

00:14:01 Speaker 2 

Of those are. 

00:14:02 Speaker 2 

In evidence, whether it’s you know, copilot for teachers, you know in interesting. 

00:14:02 Speaker 1 

Yeah. 

00:14:07 Speaker 2 

Useful ways. Their whole set of new things I think that are possible and also I’m interested in Kevin and you know. 

00:14:16 Speaker 2 

You know, we have a couple of big companies and they spend a lot of money and building these foundation models and then other people try to figure out how they can ride on top of that and in sort of an A costs an unbelievable amount of money to develop this stuff and find the talent to make things. But to start to see what’s happening at the edges with people experimenting with some of the small language models. 

00:14:24 

Yeah. 

00:14:37 Speaker 2 

Some of the more things at the edges, because I think those are going to be practical for schools and for users in all scores. So use cases that aren’t. 

00:14:44 Speaker 2 

Just the predominant one, which is. 

00:14:45 Speaker 2 

And you’re in a, you know, well resourced environment with good connectivity and and you have a good school budget to support you because that that’s only, you know, one portion of our our learning population. 

00:14:57 Speaker 1 

Yeah. Well, it’s it’s certainly as you mentioned, an exciting time in in the space. It’s it’s nice to be able to focus on on the future as as opposed to kind of. 

00:15:08 Speaker 1 

Sort of hunker down during COVID and to see all the new innovations that are coming out there. I really look forward. I want. I want you to keep your powder dry for next week, but you’ve already given us a really good taste of what should be coming up next week. So again, appreciate your time and your insights and look forward to seeing you in Miami. 

00:15:28 Speaker 2 

Thanks so much, Kevin and I look forward to reading and listening and seeing everybody in in Miami. 

00:15:36 Speaker 1 

Fantastic. 

00:15:36 

Have a great day.