Scaling Equitable AI Education with Code.org

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 share and shape the future of learning. I'm joined by Karim Meghji, Chief Product Officer at Code.org. Karim has led product and technology teams across a range of innovative industries from online media and fintech. He's been at the forefront of digital transformation for decades. At Code.org, he leads that expertise and brings it to advancing equitable access to computer science education. Welcome.

Karim Meghji:

Thank you, nice to be here.

Jeff Kavanaugh:

What inspired you to join Code.org and focus on expanding access to computer science education?

Karim Meghji:

Yeah, it's an interesting story. When I was a young person in high school myself, I was lucky enough to have a math teacher who decided to show us about how computers work to solve math problems. Honestly, I was kind of a lazy math student. I didn't love doing the longhand calculus work. And so the idea of using computers to solve those problems was really inspiring to me. And once I saw how it worked and wrote my first program to solve a math problem, I was hooked and I thought, you know, students shouldn't be lucky like I was, they should have it as part of their education and every student should have access to learning about computer science and now AI.

Jeff Kavanaugh:

It just shows you the importance about teachers and mentors and you're changing serendipity to more something that's structured. What trends in computer science education, especially with Teach.AI and the White House's AI executive order, which I'm excited about, do you think will have the biggest impact?

Karim Meghji:

Yeah, I think it's interesting when you're in the moment of such an amazing transformation that we're about to go through with AI. It doesn't feel like it necessarily in the moment. And so I do think there are a number of things that will impact how education looks in the future. I talked about one of those transformations this morning, and that is how AI tutoring will find its way into the classroom and personalize for students how it will bring a more equitable outcome for students that might not have all the tools that they might need to learn some of the things that we want them to learn to be digitally skilled for the future. So that's one trend I think that we'll see. I think teachers will have an unlock with AI as we bring AI into education. The way that we can use AI today just for efficiency gains is a great start. Let's take the things off their plate that are more rote and let's actually leverage what they do in the classroom.

Jeff Kavanaugh:

That is huge coming from family teachers. How much of the time is spent just doing those things you have to do that don't really impact the quality of the teaching.

Karim Meghji:

That's right.

Jeff Kavanaugh:

But they're administratively required.

Karim Meghji:

Yeah, like, you know, teacher wants to connect with students, learn that child's specific needs, focus on going deep with each of those children. And so I think creating space for that is another area that I see some really interesting opportunity.

Jeff Kavanaugh:

Well, I want to get the stat right. Over two million students reached through our partnership over ten years. How have collaborations like this shaped your programs and priorities?

Karim Meghji:

Yeah, we couldn't do any of the work we do without partnerships like the one we have with the Infosys Foundation. It's transformative. It enables us, frees us in many ways to think big. That's one of the things that I love about partnerships like this one. You know, the foundation has given us the space to think in ways that are audacious, that aren't constrained. And so that's probably the biggest benefit and biggest, I think, impact and outcome. We can think about things like rural kids in parts of the world where they don't have connectivity, where they don't have devices that function as well as we might have in some other parts of the world. These sorts of challenges, really hard challenges, are the challenges that, you know, when you have constraints, you don't think about those first, but when you're allowed to be unconstrained in your thinking, you can kind of visualize that in a more engaged way.

Jeff Kavanaugh:

It's exciting too with these low earth orbit satellite based internet connections so hopefully soon anywhere in the world if somebody wants to learn they can have access to that.

Karim Meghji:

Absolutely, yes.

Jeff Kavanaugh:

Of course languages too, the ability to translate into more languages with less effort, that's got to be huge as well.

Karim Meghji:

It’s huge for us. Translation is important and AI helps with that. That's another really interesting connection point where we can reach those students across those language barriers using AI. And it's not—AI can do an amazing job today of translating. So when you connect these dots, the technology unconstraining barriers to your point, internet connectivity and speaking to students in their local context, it really unlocks a lot of potential.

Jeff Kavanaugh:

Alright, that's the cool side. There's that possible dark side as well. How is Code.org helping educators integrate AI safely and responsibly through Teach.AI and programs like that?

Karim Meghji:

Yeah, so we do a couple of things. Our Teach.AI program is really built around bringing together thought leaders who have a balanced approach. They think about the opportunities and the risks and then forming policy. Policy that guides a lot of folks that are in the space of teaching and learning and using AI tools now increasingly in those contexts. So that's how we do it through Teach.AI. Through Code.org, we build products and curriculum and what we do to ensure that the products we build are safe, responsible, is we put them in the classrooms and test them. We test them with students. And we iterate and we iterate and we iterate and we don't scale these things up.

Jeff Kavanaugh:

Because you do want to be safe but not also create barriers or more administration.

Karim Meghji:

Absolutely, absolutely. But it's being, you know, having that vision, feeling unconstrained, but also grounded. I think that combination of things is how we try to approach the work we do.

Jeff Kavanaugh:

Got it. Well, being part of the Knowledge Institute, thought leadership part of Infosys, how do you think about thought leadership through facts, research, stories, data, to get your message out and to build trust and support for what you're doing?

Karim Meghji:

Yeah, a lot of that goes back to the point about community. Thought leadership is not done in a vacuum. It's coming together. It's having constructive dialogue. It's debating. It's agreeing and disagreeing. I think that's how we as a community provide thought leadership more broadly speaking. And so we ground our work in that. We don't come in with, we have the answer. We come in with curiosity and some thoughts.

Jeff Kavanaugh:

You get a position and you evolve.

Karim Meghji:

And we evolve it. And so we shape that together. And then through channels like Teach.AI with policy work, through Code.org with product work, we try and land tangible outcomes, whether that's a paper on how to bring AI into schools, to curriculum for high schoolers, teaching them about foundational AI concepts.

Jeff Kavanaugh:

And measuring outcomes, I'm sure, as your folks go through that. That's right. How do you gather and apply this—speaking of that—apply this feedback from teachers and students to improve your offerings? You're Chief Product Officer, so what does that mean to incorporate back into your products?

Karim Meghji:

You know the product is never done. That's one of the things I think about as a Chief Product Officer. You can always make it better. It goes back to the earlier thing we talked about Jeff, which is iteration. You know when you're at stages like right now where AI is transforming so many things, it's not about a constant outcome. It's about finding the next place that you can go to make the product better. And so again it's hours in the classroom, hours with teachers, iterating, iterating, iterating, and really learning what those outcomes look like, measuring them to your point and then building on that.

Jeff Kavanaugh:

Well, looking ahead for the next five years, let's say, what are your top priorities and how do you plan to get there?

Karim Meghji:

So starting kind of at the top, it's really framing what is AI literacy and foundational AI knowledge need to look like for that student who will be a part of this digital future that we see in 10 years, 20 years. We're really thinking hard about durable skills. What does that mean in a world with AI? We're really thinking about what does computer science look like in the future? Something that's on a lot of people's minds. So it begins with just thinking and understanding the concepts that we need to teach students.

We do see a role for AI software in our products and so thinking about an equitable, accessible AI tutor that can really help with some of the things we've already touched on. On the student side, thinking about what are the teacher supports that AI can provide through what we call the AI teaching assistant, which is probably not surprising. And then how do we manifest some of the earlier points I touched on in AI curriculum? Curriculum that provides that student that piece of knowledge that ultimately when they get into the workforce and they're in an AI native environment they can work with those systems and those tools in a responsible, ethical, productive way—not just an efficient way.

Jeff Kavanaugh:

Great, well I’ll let you get back to the main Foundation events, so I appreciate you joining. Thank you so much for all your contribution, your time today, and as we like to say at Knowledge Institute, keep learning and keep sharing.