Join us on Thursday, April 30th

12pm-1:30pm

University at Buffalo South Campus, Foster 135

with
Dr. Krista Glazewski, Executive Director, the Friday Institute. Associate dean for translational research in the College of Education, North Carolina State University

Krista Glazewski, PhD, As executive director of the Friday Institute and associate dean for translational research in the College of Education, Dr. Glazewski provides executive leadership and management of the Friday Institute, including collaborations and partnerships. Dr. Glazewski provides executive leadership for the Friday Institute, guiding a team of more than 70 staff, affiliated faculty, and students in advancing its mission to innovate at the intersection of research, practice, and policy. The Institute develops evidence-based solutions to pressing challenges in K–12 education, builds educator and leader capacity, and translates research into scalable practices and technologies. Spanning multiple regions in the U.S., Dr. Glazewski’s work investigates how and under what conditions teachers might adopt and adapt new practices, with a focus on complex problem solving with collaborative inquiry supported by foundational and emergent technologies. Her work has resulted in numerous contributions, over 70 publications and $14 million in state, federal, and foundation funding. She is a former middle school teacher originally from New Mexico.

Forum Abstract: Artificial intelligence in education is often framed in terms of automation. This work advances an alternative vision in which AI is designed to expand human intelligence within collaborative learning environments. Agency is positioned as central across learners, teachers, and researchers, with AI supporting sensemaking, instructional orchestration, and analytic insight while humans retain interpretive control. Drawing on classroom-based designs, including narrative-centered environments and teacher-in-the-loop authoring tools, the paper illustrates how AI can make student thinking visible, coordinate collaborative reasoning, and inform real-time instructional decisions. These systems exemplify intelligence augmentation by amplifying human judgment, creativity, and collective inquiry. The argument foregrounds the careful distribution of expertise across people and tools. It outlines design principles for learning ecosystems that sustain ambitious, inquiry-driven practice while strengthening human agency in the presence of increasingly capable AI systems.

Thursday, February 12th

12pm-1:30pm

University at Buffalo North Campus

with
Jeremy Roschelle, Executive Director, Learning Sciences Research, Digital Promise

Jeremy Roschelle, PhD, is a nationally recognized leader in the learning sciences and education innovation. Digital Promise is a non-profit organization originally authorized by Congress to accelerate innovation in education, with a long history of bridging research, practice, and policy. Over the past decade, Digital Promise has played a central role in shaping the national AI-in-Education landscape, including major initiatives such as the NSF-funded ENGAGE AI (a sister institute to our AI4ExceptionalEd) and the IES-funded U-GAIN Center (a sister center to CELaRAI). Most recently, Digital Promise has been entrusted by seven major philanthropic foundations to steward nearly $100 million in national K–12 AI education infrastructure investments, with the first round of competitive RFPs expected to launch in late January or early February. This work places Digital Promise—and Jeremy personally—at the center of national conversations about how researchers, educators, and institutions of higher education can collaboratively lead AI innovation in education.

Abstract: Today’s AI can make everything seem possible, and yet nothing in education yields to superficial or easy solutions. At Digital Promise, our team works the intersections among researchers, educators, technologists, and policy makers—and this makes us think deeply about challenges that resonate deeply across perspectives. With the limited time and resources available, how can we identify challenges worthy of the partnerships? In this talk, I share thoughts on some of these challenges, each illustrated with some of our recent or ongoing work — and invite participants to think together about criteria and processes for identifying worthy challenges. We will discuss topics such as: How can we work in partnership with educators in ways that build transparency and trust? How can communities develop public goods that raise the bar for depth of learning science principles incorporated into AI-enabled services? And how can research-based improvements in AI in education get to scale?

Graduate Student & Postdoc Session

2:00–3:00 p.m.Following the forum, Dr. Roschelle will host a dedicated session for UB graduate students and postdocs focused on alternative career pathways in the education innovation ecosystem. The discussion will explore roles in non-profit research institutes, learning sciences organizations, philanthropic initiatives, and public-interest technology—alongside how these paths compare to academic careers. If you’re thinking about the broader landscape of learning sciences careers, this is a unique opportunity to hear firsthand from someone working at the national level. Open to all UB graduate students and postdocs.

Our August 2026 Monthly Talk Speaker will be announced soon. Register now to join us on Friday, August 14th at 12pmET!

Our July 2026 Monthly Talk Speaker will be announced soon. Register now to join us on Friday, July 10th at 12pmET!

Erin Mote is the CEO and Founder of InnovateEDU. In this role, Erin leads the organization and its major projects, including its policy and strategy portfolio.  She leads the organization’s
work on creating uncommon alliances to create systems change – in special education, talent development, artificial intelligence, and data modernization.  An enterprise architect, she created, alongside her team, two of InnovateEDU’s signature technology products
–  Cortex, a next-generation personalized learning platform, and Landing Zone – a cutting-edge infrastructure as a service data product. 

Session Information: Building Responsible AI Ecosystems
As AI moves from a novelty to a foundational layer of educational infrastructure, the challenge shifts from simple adoption to the creation of a responsible ecosystem. This session, led by Erin Mote, explores the critical intersection of education, policy, and technology through the lens of the “S.A.F.E. by Design” framework.

Attendees will examine how to move beyond consumer-grade tools toward purpose-built educational AI that prioritizes the Science of Learning and Development (SoLD). We will discuss the “policy stack”—integrating state standards, data interoperability, and AI literacy—to ensure technology serves as a bridge to equity. Participants will leave with actionable strategies for fostering uncommon alliances between practitioners, policymakers, and industry leaders to build a trusted, evidence-based future for all learners.

Effective AI for Learning: Highlights of my 40 Year Journey

AI in Education is a 40+ year old field with lots of lessons learned. The importance of combining “technique” with “technology” is one of them. As researchers and developers, we will produce more effective AI for learning by drawing both on the learning sciences and on iterative engineering methods that use data to drive continual improvement. I illustrate these lessons with some of our successful AI applications, particularly in mathematics and training of human tutors. One surprising general result of our work is the discovery of an astonishing regularity in student learning rate. This result suggests that with the right kind of favorable learning conditions any learner willing to invest the needed effort can succeed. Helping students to invest such effort is a challenge we are addressing through an on-going project combining human and AI-based tutoring.

Kenneth R. Koedinger is the Hillman University Professor of Computer Science at Carnegie Mellon University, with joint appointments in Human-Computer Interaction and Psychology. Dr. Koedinger’s multidisciplinary background—including an M.S. in Computer Science, a Ph.D. in Cognitive Psychology, and experience teaching in urban high schools—underpins his research goal of understanding human learning to create technologies that dramatically increase student achievement.

He is the director of LearnLab, the scientific arm of CMU’s Simon Initiative, which leverages computational approaches to identify the conditions that cause robust learning. His research has produced fundamental cognitive science results on student thinking and has pioneered new principles for educational software design. He was a co-founder of Carnegie Learning, Inc., which has brought AI-driven “Cognitive Tutor” courses to millions of students since its inception in 1998.

Currently, Dr. Koedinger leads the PLUS project, providing hybrid human-AI tutoring to middle school math students nationwide, and oversees major data-sharing infrastructures including DataShop and LearnSphere. With over 250 peer-reviewed publications and more than 45 grants, his work has repeatedly demonstrated significant learning gains, such as doubling the rate of algebra proficiency in a single school year.

Dr. Koedinger is a Fellow of the Association for Computing Machinery (ACM), the Cognitive Science Society, and the Association for Psychological Science.

Project LISTEN’s Reading Tutor listened to children read aloud and helped them learn to read. It displayed text on a computer screen, used automatic speech recognition to help analyze oral reading, and responded with spoken and graphical assistance modeled after expert reading teachers but adapted to affordances and limitations of the technology. The Reading Tutor logged its interactions in detail to a database that we mined to assess students’ performance, model their learning, and harvest within-subject embedded experiments to compare alternative tutorial actions. This talk will illustrate a few of the lessons we learned about children, reading, speech technology, intelligent tutors, educational data mining, and doing AIED research in schools.

Jack Mostow, Emeritus Research Professor of Robotics, Machine Learning, Language Technologies, and Human-Computer Interaction, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213-3890, mostow@cs.cmu.edu

Jack Mostow founded Project LISTEN to develop an automated Reading Tutor that listens to children read aloud. It won AAAI94’s Outstanding Paper Award, a U.S. patent, inclusion in NSF’s Nifty Fifty, and the Allen Newell Medal of Research Excellence. He subsequently led the RoboTutor team (robotutor.org), a $1M Finalist in the $15M Global Learning XPRIZE competition to develop an open-source Android tablet app to enable children in developing countries to learn basic literacy and numeracy without relying on a literate adult.

Prof. Mostow earned his A.B. cum laude in Applied Mathematics from Harvard University and his PhD in Computer Science from Carnegie Mellon University. He served as President of the International Artificial Intelligence in Education Society and has authored over 200 publications in artificial intelligence, educational data mining, intelligent tutors, user modeling, machine learning, and speech and language technologies.

Join us for an exciting talk on how AI and speech technology can support learning to read, bridging scientific research and real-world products. Dr. Helmer Strik and his team have been involved in several national and international projects on Literacy and Reading, both fundamental and applied research (see https://www.helmer-strik.nl/projects/). Besides interesting scientific results, two of these projects have led to products of publishers that are now used by many learners. Dr. Strik will present an overview of his team’s research, with a selection of results and insights obtained, focusing on how AI and speech technology can be used to improve Literacy and Reading skills.

Dr. Helmer Strik

Helmer Strik received his MA and PhD in Physics from Radboud University, where he later became Associate Professor in Speech Science and Technology. His research activities address AI-based automatic speech recognition (ASR), speech technology, computer assisted language learning (CALL), e-Learning, and e-Health. He has published over 350 refereed papers, and has obtained various national and international grants for his research (see https://www.helmer-strik.nl/projects/).
He is co-founder of two spin-off companies, and is Chair of the ‘International Speech Communication Association’ (ISCA) ‘Special Interest Group’ (SIG) on ‘Speech and Language Technology in Education’ (SLaTE; see: https://www.helmer-strik.nl/slate/).

What does it mean for young learners to become literate in a world where they can write with AI? This interactive session invites participants to rethink foundational literacy practices through the lens of AI-supported reading and writing. Drawing from the Compose with AI and Compose with AI Foundations projects, Dr. Bruner shares classroom examples and practical strategies that show how even young children can question, interpret, and build upon AI-generated text. Participants will leave with ideas for integrating AI literacy, authorship, and ethics into early literacy and content area instruction.

Dr. Lori Bruner
Assistant Professor in the Department of Literacy Teaching & Learning at the University at Albany.

Dr. Bruner’s research examines how digital media supports early literacy development, with a focus on children’s vocabulary learning and the role of AI in classrooms. Her recent projects, funded by the International Literacy Association and the National Science Foundation, explore vocabulary in digital texts and the design of AI-supported writing tools for young learners.