#1261: Integrating the Learning Sciences into Technical Infrastructure for the Future of AI in Education

As artificial intelligence becomes increasingly influential in education, attention must extend beyond AI applications to the technical infrastructure on which those applications depend. Datasets, models, benchmarks, and governance mechanisms are already shaping the future of educational AI, yet these foundational resources are often developed without sufficient engagement with the learning sciences. This workshop explores how learning scientists can contribute to the design of such infrastructural public goods in ways that better reflect what is known about learning, while also addressing concerns related to safety, privacy, ethics, and equity. By bringing together researchers working at the intersection of AI and learning, the workshop seeks to identify promising lines of work, articulate key tensions, and stimulate future collaboration on learning-science-informed AI infrastructure.