Askwith Education Forum Brings AI Theory Into Classroom Practice

The impact of artificial intelligence on education remained at the forefront of the latest Askwith Education Forum at the Harvard Graduate School of Education on Wednesday.

Building on an AI-themed Askwith from September in which Professor Howard Gardner discussed how cognition and education have shifted amid the rise of generative AI and large language models (LLM), Wednesday evening in Askwith Hall featured educators sharing how AI has made a direct impact in the classroom.

The event was an opportunity to move the discussion about AI from the conceptual to the concrete, featuring a conversation moderated by Associate Professor Ying Xu. She led three panelists through concrete ways they’ve seen AI change classroom learning. The panelists included Keith Parker, the superintendent of Elizabeth City-Pasquotank Public Schools in North Carolina; Yenda Prado, Ed.M.’05, a learning sciences research analyst at Digital Promise; and Kedaar Sridhar, Ed.M.’25, co-founder and chief product and tech officer at M7E AI.

“This is exactly the kind of inquiry that is core to our mission here at HGSE,” said Dean Nonie Lesaux, introducing the evening’s panel. “Because when we talk about AI in schools, it’s really important to remember we’re not just talking about technology. We’re talking about opportunities for students, for educators, and for communities. We’re talking about ensuring that every learner benefits from innovation.”

Parker, whose district is piloting a microschool that uses AI in a variety of ways, detailed how the ability to quickly test new ideas and technologies has had a positive impact on its classrooms. He described using ChatGPT to scan classroom text to instantly learn its Lexile level and make sure students were being taught with grade level-appropriate materials.

“That’s a tool that we’ve not had,” said Parker. “Being able to take lesson plans and texts and modify them, it simplifies the role of educator and makes life easier with that tool. There’s so much practical application for that.”

Sridhar recently started a company with HGSE classmate Abdirahman Guleed, Ed.M.’25, that builds tools to modify the language of math word problems to help make them more accessible across cultures and languages. He advocated for how AI can be an “equalizer” for a variety of learners.

“The goal in general for us is how can we improve the curriculum before it even hits the classroom?” Sridhar asked. “How do we best serve the upstream before districts have to purchase curriculum, before teachers have to teach it, and before students have to engage in the struggles of learning in the classroom there?”

As panelists detailed how their work in the field impacts learning, big questions about the future weighed heavy in the discussion. While the emergence of large language models has shifted the way students learn so quickly, there’s a lot researchers and teachers are adjusting to as that tech develops.

“We’re in this era of knowing a whole lot and knowing nothing,” said Prado. “And how can those two things be true at the same time?”

Prado suggested educators working with AI help create a knowledge base of how these tools can actually impact classroom learning to help develop new strategies that work with different kinds of learners.

“How can we use the things we know AI is good at — pattern recognition, statistical inferencing — how can we use those assets and couple them with what we know works from the learning sciences in terms of how students learn?” asked Prado. “And what does that look like?”

You can watch the last Askwith of the fall semester in the video above.