The University of Florida Literacy Institute (UFLI), in partnership with Project Read AI, recently received a $2,000,000 grant to support reading instruction and proficiency. Funded by the U.S. Department of Education’s ETechM² Program, the five-year UFLIAI: Advancing Foundational Literacy with Artificial Intelligence study will develop an AI-enhanced learning analytic tool integrated into UFLI Foundations. Matthew Burns, Ph.D., a Fien Professor of Special Education, will lead the project. It will be supported by co-PIs Holly Lane, Ph.D., the director of the University of Florida Literacy Institute and Valentina Contesse, Ph.D., clinical assistant professor of special education and co-author of UFLI Foundations.
Reading ability affects a student’s performance in every other aspect of school. UFLI Foundations was developed to address this issue. Based on decades of research and the science of reading, the program has been adopted by teachers, schools and districts worldwide. When implemented as intended, the program results in students demonstrating significant gains in early literacy outcomes. However, many teachers struggle to make data-based decisions about how to best individualize small-group instruction for students. Time and resources are often cited as factors that impede teachers’ ability to individualize instruction effectively.
The UFLI AI project will radically change this limitation by developing and refining an AI-enhanced learning analytics tool that teachers can use to provide small-group supplemental support and individualized, AI-driven support. Central to this project is a web-based decision-support portal on Project Read that will use learning analytics to streamline teachers’ instructional planning.
Teachers of kindergarten through second grade students will enter weekly spelling test data into the portal, which will automatically analyze performance and generate tailored recommendations for small-group instruction.
“UFLI AI will offer a cohesive system that empowers teachers to make timely, data-informed instructional decisions while providing students with immediate, individualized practice opportunities,” said Burns.
“In our initial pilot of this system, teachers found the portal to be a real game changer. We are excited about what’s possible as we further refine the tool,” added Lane.
As teachers implement differentiated small-group instruction, students will be able to independently engage with the Project Read AI Tutor, which will use data from the portal to individualize support. This practice tool listens to students read decodable text aloud and delivers real-time feedback. It will guide students to correct misread words, reinforcing letter–sound relationships and building reading fluency.
UFLI AI will directly benefit all students learning to read, including those with specific learning disabilities (such as dyslexia), speech-language impairments, or other developmental delays that affect literacy acquisition. To ensure the tools are useful to students and teachers, UF researchers will work with teachers across multiple partner schools. “Educators serving students with disabilities will be engaged as co-designers throughout the process,” said Burns. “They will be asked to provide iterative feedback on instructional content, user interfaces, accessibility features and implementation supports.”
By embedding evidence-based instructional decision-making processes into an intuitive, user-friendly platform and combining them with innovative AI-enabled practice, UFLI AI will be scalable, sustainable, and positioned to improve foundational reading outcomes.
Want to learn more about UFLI? Be sure to visit their website.
Matthew Burns, Ph.D.
Valentina Contesse, Ph.D.
Holly Lane, Ph.D.