AHA Seminar: Osbert Bastani | Perils and Promise of Generative AI in Education
Register with organizerWhat to look forward to
Osbert Bastani, a University of Pennsylvania associate professor who leads the trustml@Penn research group and holds a Stanford Ph.D., will share findings from large-scale studies on AI and student learning. He’ll explore how unguided AI use and self-regulated help-seeking can undermine learning, and how proactive, personalized AI tutoring and critical-engagement training can improve it.
Inside the event
Join us for our online seminar series event hosted by MIT Media Lab's Advancing Humans with AI (AHA) research program! This event features Osbert Bastani, Associate Professor at the University of Pennsylvania in the Department of Computer and Information Science. Talks are all recorded and made available afterwards on YouTube, linked on each event webpage. Abstract The rapid integration of AI into educational settings presents both opportunities and challenges. In this session of AHA Seminar Series, Osbert will discuss findings from three large-scale field studies investigating the impact of AI on student learning. After finding that unfettered access to ChatGPT negatively affected students’ short-term learning outcomes, his team has focused on understanding longer-term effects by examining learning in chess academies. Contrary to the popular strategy of promoting student agency, these findings show that self-regulated learning (where students decide when to request AI help) can substantially harm learning by diminishing engagement and motivation. Osbert will also show how integrating GenAI tutors with proactive personalization can improve student learning. Taken together, his studies suggest that while giving students unguided AI tools can be detrimental, targeted interventions that train students to engage critically with AI can be beneficial. Speaker Bio Osbert Bastani is an Associate Professor in the Department of Computer and Information Science at the University of Pennsylvania, where he leads the "trustml@Penn" research group on trustworthy machine learning. He is a member of the ASSET, Schlein, PRECISE, and PRiML centers, as well as PLClub. He completed his Ph.D. at Stanford, advised by Alex Aiken, and then spent a year as a postdoc at MIT working with Armando Solar-Lezama.
A seat from anywhere.
Register with the organizer for joining instructions and the event link.
Discovered through Cerebral Valley. Details come from the event listing; registration is managed by the organizer.
Last checked: Oct 13, 2026. Confirm current availability before making plans.