Empowering Diverse Learners in the Age of AI

Jeff Kavanaugh:

I'm Jeff Kavanaugh, Global Head of the Infosys Knowledge Institute, and we're here at Infosys Foundation USA Crossroads 2025, where leaders across education, technology, and social impact have come together to shape the future of learning. I'm joined by Dr. Maya Israel, Associate Professor at the University of Florida and Director of the CS Everyone Center, and Andrew Bennett, a PhD candidate and Graduate Research Assistant at the Center. Together they're working to make CS, computer science, and AI education more inclusive for all students. Thanks again for joining us.

Maya Israel:

Thank you.

Andrew Bennett:

You bet.

Jeff Kavanaugh:

First question for you, Maya. What's one of your most impactful projects so far?

Maya Israel:

So one of my most impactful projects so far is a project that brings together school districts and the Computer Science Teachers Association together with us to solve some of the major problems around how to support kids with disabilities in computer science and AI education. And the way that we are coming together is by actually building a community called a Networked Improvement Community where we can think about the major challenges that school districts face around computer science education and then try to address them together. For example, is it an issue of the technology not being accessible? Is it an issue of teachers not having the right professional development? Is it an issue of like broad systems infrastructure issues as well?

Jeff Kavanaugh:

And is that locally, counties around you, in Florida, nationally, what's your scope?

Maya Israel:

Yeah, so the scope is national and we started actually with a small grant from Infosys Foundation USA with Broward County Public Schools back before the pandemic. And so Broward County has been a major, major partner with us.

Jeff Kavanaugh:

How has the Infosys Foundation USA supported your work beyond that initial seed funding?

Maya Israel:

So Infosys Foundation USA, I think they did a couple of things. The first is obviously this seed funding that started the work back in 2018. Since then, there have been a couple of amazing initiatives. One of them is the Pathfinders Institute. So every year we come together, we bring in teachers from across the country to learn about the inclusive pedagogies to support kids with disabilities. So we love Pathfinders and our cohorts of teachers who've been part of Pathfinders are able to take the information they learn from us and actually apply it in their classrooms. The second opportunity that we've had is that we've been working with Arizona State University and the University of Florida and the local school district in Arizona to be able to do some really intensive professional development with the teachers and think about in a school district like Avondale Unified Elementary School District that actually has a pretty good reputation for inclusion of kids with disabilities but that hasn't necessarily made it to computer science education, what can we actually do to increase that access? And so actually that particular project is bringing together computer science teachers, special education teachers, even paraeducators for professional learning.

Jeff Kavanaugh:

Well, Andrew, over to you. What tools or strategies are helping teachers support diverse learners?

Andrew Bennett:

Frameworks that support a wide range of variabilities like universal design for learning and high leverage practices to support learning, engagement, and accessibility. From an accessibility perspective, use the perceivable, operable, understandable, and robust principles, meaning physically accessible, being navigable and usable, clear and consistent directions in navigation, and being able to work with a variety of technologies, including assistive technologies. When we talk to in-service teachers and those in teacher training programs, we also emphasize the usefulness of assistive technologies like screen readers, modified keyboards, and so on in supporting students with disabilities specifically.

Jeff Kavanaugh:

It's amazing how this both universal design and also the tools at the moment of truth to assist the learning. Anything you wanted to add to that?

Maya Israel:

Yeah, I think there's a combination of can the tools be designed with access in mind? And then on top of it, can the teachers use their own understanding of their learners to be able to make decisions about which tools are the most appropriate for their students? And so one of the things I love about Andrew's work is that you're working with teachers and accessibility is not an all or nothing situation. It’s really about is the tool appropriate for the learner. And so in the context of computer science and AI and maker education, it gets really complicated because we want kids to engage in really authentic ways. It's not as simple as can a screen reader be used with it. It's about can a student authentically create with it.

Jeff Kavanaugh:

Authentically tailoring at scale, you know.

Maya Israel:

Yes.

Jeff Kavanaugh:

Andrew, back over to you. I'd imagine that it's not a perfect world out there at all and you’ve got a lot of things to overcome in getting the word out. What systemic changes are needed to scale inclusive computer science and AI?

Andrew Bennett:

High quality professional development, especially in long term professional learning communities has been something we have seen as being effective, especially in improving teachers’ knowledge and understanding of practices and tools needed for students to succeed. Accessible technologies, tools and materials backed by effective teaching and learning strategies are also useful because they give the students the chance to participate, understand, and engage with the material. Finally, recognizing the importance of computer science for all students, including students with disabilities, is crucial for scaling. Such recognition concentrates on the importance of technology in today's economic landscape, even in jobs beyond computer science, and, moreover, recognizes that computer science is rapidly becoming an important part of society. Alongside this recognition is the importance of taking a strengths-based approach of what students can do and accomplish and then examining ways in which to maximize those strengths and eliminate or diminish barriers.

Jeff Kavanaugh:

Well, you said a lot there. Anything you wanted to add?

Maya Israel:

Man. I guess the thing that you said, Andrew, that really spoke to me is that it's not just about the classroom experience. It's about preparation for future careers. And so the why we're doing this is to make sure that kids feel included in their education, but it's also about making sure that they are prepared for the jobs of today, the jobs of tomorrow, where technology is just part of everyday life and is part of many careers.

Jeff Kavanaugh:

And if there are those 1.9 million job shortages like in manufacturing and other places, let's go find these folks to go fill those jobs.

Maya Israel:

Right, and I think that's a really important piece because we often have misconceptions about who belongs in tech-supported fields. And those misconceptions aren't based on data. They're not based on the potential that kids have. They're based on what people perceive people with disabilities can and can't do. And so the work that we do is a piece of it, right? So if teachers have the skills and the tools to be able to provide experiences to kids around computing in AI, that's a start, right? The other piece is are the industries open? Are they accepting? Are they inclusive? So that when students are prepared and have the skills, they actually can go and be competitive for those positions.

Jeff Kavanaugh:

I mentioned acronyms before. There is one. P-O-U-R. POUR. Andrew, how are you using POUR principles in your research? And what is POUR?

Andrew Bennett:

POUR principles in two ways. First, we are doing professional development with a professional learning community. We are promoting the use of POUR as a way for teachers to be able to evaluate and modify the technologies and materials they use to teach computing. We have already learned a lot from the creativity and effort the teachers in the PLC are putting into their projects. Especially the decisions they make about using and modifying materials, especially based on their POUR analysis. That allows us to generate feedback for technology creators they can use to improve their product. Secondly, we are starting to think about how we can combine POUR with universal design for learning and high leverage practices to support student learning.

Jeff Kavanaugh:

Got it. And POUR stands for?

Maya Israel:

It stands for, so POUR is a framework that can be used to evaluate the accessibility of tools. And often accessibility is so complicated. And so as a teacher, it's really hard to say like, is this compliant with this particular piece of specific standards? What POUR is, is a large framework where you can think about, is something perceivable? Can you actually perceive the information on the screen? Is it operable, right? Can you navigate it with whatever tools you have? Is it understandable? Can you make sense of what is happening from a cognitive perspective? And is it robust? Can it be used with additional supports and infrastructures? And so even though accessibility is really complicated, the framework itself is one that you found teachers are able to understand.

Jeff Kavanaugh:

Broadly speaking, Maya, why is inclusion essential in computer science education?

Maya Israel:

So, inclusion is essential in computer science education. It's essential broadly. So, computer science is not unique.

Jeff Kavanaugh:

But what specific about it really brings it out though?

Maya Israel:

Yeah, so I think what's specific is if we think about the design of tools, if we don't have the broadest range of learners who are part of the design process, then we're missing a huge opportunity in who the end users are. So, if we don't have developers who have disabilities part of development teams, then we don't know what we don't know, right? So, as we're designing tools for everyone, it's critical for students with disabilities, adults with disabilities to be part of the decision-making process. At the same time, it's also, I think, an important aspect of making sure that opportunities are out there. So, you talked about careers a little bit ago and the fact that there are jobs out there that are very tech forward and that companies are looking for highly skilled employees. And so, if we don't actually think about making sure that all kids, including kids with disabilities, are getting that education, then we're completely leaving them out of those opportunities.

Jeff Kavanaugh:

Well, thinking about the future, what's next for your work at the CSEveryone Center? And maybe, Andrew, you could start with that.

Andrew Bennett:

I'm the primary investigator on a $450,000 two-year grant from Google to apply the POUR principle to the Google Blockly programming language. We are looking forward to working with the Blockly team and our professional learning communities on that, especially on decision-making and learning strategies. Long-term goals beyond that are to scale up our findings from our work with the PLC, examine accessibility decisions made by district curriculum directors and other administrators as well as the preferences of students with disabilities, and further explore ways to promote accessibility in robotics and AI education including human influence and vice versa. I am also very interested in computer science teacher decision making, especially about accessibility, which is the subject of my dissertation on identity development. I like engaging with both reservists and in-service teachers, so that's always part of my work too.

Jeff Kavanaugh:

Awesome. What's next?

Maya Israel:

So we've been thinking a lot about how to scale the work that we do. This work is very intensive in terms of direct support to teachers. And that's not scalable. So what we're now doing is we're developing a couple of different models around scalability. One is a train the trainer model. So people will be able to come to the University of Florida or we'll be able to come to them and provide leadership teams with the knowledge so that they can provide this professional development within their school districts. So that's a really exciting thing that's coming up very soon. And the other part is that we have teachers around the world who even in the best-case scenarios don't have access to this level of professional development. So even if we had a train the trainer model, they may not have access to that. So we're developing a series of micro credentials for individual teachers across the world who want to learn about accessibility or universal design for learning or high leverage practices in the context of computer science and AI education. So those are in development. The first one is actually going to be launching this summer. So we're really, really proud of it. But the what's next is really about scaling all of the knowledge and information that we've been working on over the last many, many years.

Jeff Kavanaugh:

Thought leadership sometimes is thought to be insights and things like that. How valuable is it with facts, research, stories to get your message out either to partners, funders, or just the people you're trying to reach?

Maya Israel:

Yeah, thank you for that question. I think it's critical. One of the things that I find in our work is that we're so busy in the work that we don't communicate about what we're doing. And it's really important to get the word out, A, about the work itself, the success that our teachers and our students are finding. But it's also important kind of in terms of the impact of what we're doing to be able to talk with other thought leaders and funders, because I think that there's so much potential. And if we just keep our head down and we do the work, nobody will find out about it. And so we're really grateful to be able to highlight this work because it's critical. We feel really, really good about it and we're really proud of it.

Jeff Kavanaugh:

Well, I can tell, and we're so proud to have you with us as well. Thanks for joining us because your work in promoting computer science and AI education is so important. As we like to say at the Knowledge Institute, keep learning and keep sharing.