Penn GSE Invests $26 Million in Educational AI

The graduate school partners with Digital Promise to develop publicly accessible AI tools for K-12 schools.

Feb. 5, 2026 at 11:55pm

Got story updates? Submit your updates here. ›

Penn's Graduate School of Education has announced a $26 million partnership with Digital Promise and other organizations to develop publicly accessible generative AI tools for use in K-12 education. The initiative, called the K-12 AI Infrastructure Program, will provide grants to build AI products aimed at improving formative assessment and supporting instructional leadership.

Why it matters

This investment highlights Penn GSE's leadership in promoting ethical and effective use of AI in education. The program aims to create open datasets and infrastructure to improve AI's ability to account for factors like learner variability and teaching principles, bridging the gap between technology and evidence-based learning.

The details

The K-12 AI Infrastructure Program will be led by Catalyst @ Penn GSE, which is joining forces with the Massive Data Institute at Georgetown University, Learning Data Insights, and DrivenData as core partners. The initiative will focus on 'formative assessment' - using feedback from students and educators to understand frustrations with existing tools and develop better AI assistants and partners for instruction.

  • The program was announced on January 29, 2026.
  • Penn GSE previously collaborated with Digital Promise in 2024 to launch a certificate course on data science methods for digital learning platforms.

The players

Catalyst @ Penn GSE

A program at the University of Pennsylvania's Graduate School of Education that is leading the K-12 AI Infrastructure Program.

Digital Promise

A nonprofit organization that is a core partner in the K-12 AI Infrastructure Program, which aims to develop publicly accessible generative AI tools for use in K-12 education.

Jeremy Roschelle

The executive director of the K-12 AI Infrastructure Program.

Sijie Mei

A second-year graduate student at Penn GSE who believes the grant is crucial because it focuses on infrastructure to bridge the gap between technology and evidence-based learning.

Annie Yang

A first-year graduate student at Penn GSE who built an AI-powered platform supporting families of autistic children, and sees the grant as an important step in expanding opportunities for GSE students to build and test AI models.

Got photos? Submit your photos here. ›

What they’re saying

“There isn't an established way that people collect data sets in education for AI or build benchmarks or tune models — it's all pretty new. We really needed to build a first class set of partners to do that work.”

— Jeremy Roschelle, Executive Director, K-12 AI Infrastructure Program (The Daily Pennsylvanian)

“If we're going to get AI to really support the learner, there needs to be data that's appropriate to that, and I think we're going to chip away at that problem.”

— Jeremy Roschelle, Executive Director, K-12 AI Infrastructure Program (The Daily Pennsylvanian)

“AI is powerful, but it doesn't really always understand pedagogy. For education, we need to be very cautious about the data we use to train. We need to use [data that] has been verified or validated by science, and unbiased tools.”

— Sijie Mei, Second-year GSE Graduate Student (The Daily Pennsylvanian)

What’s next

The K-12 AI Infrastructure Program intends to improve AI infrastructure by creating free and legally unrestricted open datasets used to train machine learning models, with the goal of improving AI's ability to account for factors like learner variability and teaching principles.

The takeaway

This investment by Penn GSE demonstrates the school's leadership in promoting the ethical and effective use of AI in education. By partnering with Digital Promise and other organizations, Penn GSE is working to bridge the gap between technology and evidence-based learning, with the ultimate aim of enhancing educational outcomes for K-12 students.