NSF Renews $20 Million Grant for EngageAI to Transform K-12 STEM Education

The National Science Foundation has renewed a $20 million award for the AI Institute for Engaged Learning, known as EngageAI, extending the project's operations through 2031. This funding ensures the continued development of artificial intelligence tools designed to modernize and improve K-12 STEM education.

The grant supports a broad collaborative effort involving NC State University, UNC-Chapel Hill, Vanderbilt University, Indiana University, and the nonprofit Digital Promise. A primary focus of the renewed funding is the creation of narrative-centered learning environments. As part of these efforts, the team has developed EngageVP, a multimodal video analysis framework. This tool allows researchers to analyze video, speech, and gesture data to better understand how students engage in collaborative problem-solving within the classroom.

The Evolution of Engaged Learning

Launched in 2021, the EngageAI Institute was established to study how AI can support collaborative learning in science, technology, engineering, and mathematics. The project leverages core AI research expertise, particularly from UNC-Chapel Hill, to move beyond traditional rote memorization. By focusing on narrative-centered environments, the institute seeks to integrate storytelling and interactive elements into the curriculum, making complex STEM concepts more accessible to younger students.

Democratizing STEM Instruction

This initiative aims to democratize individualized instruction. In many K-12 settings, teachers struggle to provide one-on-one support to every student in a large classroom. The project intends to solve this by creating AI-powered lessons that adapt to a student's specific progress over time. By utilizing world model AI technologies for narrative generation, the institute hopes to provide educators with more dynamic tools to enhance classroom engagement and tailor the learning experience to the needs of the individual.

Future Outlook

As the institute enters this five-year renewal phase, the focus will shift toward scaling these adaptive tools for wider classroom use. While the EngageVP framework has already demonstrated the ability to track student interaction, the next stage of development will center on how these insights can be translated into real-time instructional support. Observers will be watching to see how these narrative-driven AI models integrate into existing state curricula and whether the personalized support leads to measurable improvements in STEM proficiency across diverse student populations.