Dan Restuccia

Next phase of the higher ed AI integration revolution = empirical evidence, empirical evidence, empirical evidence. I read this morning's Faculty Focus article - Designing the 2026 Classroom: Emerging Learning Trends in an AI-Powered Education System. The author included this statement: "An AIPRM report of U.S. students, for instance, found a 62% increase in test scores among those using AI-powered instruction systems, attributed to the technology’s ability to identify and address knowledge gaps before they develop into larger challenges." Clicked on the link to the report, and this is a study conducted by an adaptive-learning company that compared student test scores of those using their program to those who didn't... Across the remainder of the article, the majority of the "pro-AI" evidence came from multiple EdTech companies and leaders, while the critical evidence (one single source - Education Week...) came from what I would consider a less biased source. In the business world, ROI has not met expectations when it comes to AI investment (yet). Educators - where we typically measure our ROI in student learning outcomes achievement, how are we measuring success and failure at scale? This year, my goal is to develop a valid assessment instrument/process to determine where AI is succeeding and falling short. Has anyone else been using or created such a tool? I would especially love to partner across institutions on a wide-scale study! I believe we're going to see more and more empirical claims of success from EdTech. I'd much rather see balanced evidence coming directly from higher ed.

7

Three essential points of consideration from recent *Harvard commentary* on the state of U.S. math education (article linked below): 1. “Math is cumulative: If you don’t understand fractions when they’re introduced in third and fourth grade, you’re going to have a lot of trouble when you get to algebra.” 2. “The more modern curriculum materials are harder to use.” 3. “To teach multiple procedures, to engage students in conversations in which ones are better and why … it feels a bit more attainable.” ⚡Each point is a window into a *system problem*. But together, they also point to a system solution, which addresses student learning *and* teacher learning through coherent tools and routines.⚡ That’s the promise of OGAP 🧠 In the context of High-Quality Instructional Materials (HQIM) and High-Quality Instructional Experiences (HQIX), OGAP doesn’t compete--at all--with curriculum, but instead it #empowersteachers to deliver #HQIX that makes curriculum work: ▪️ Anchors instruction in learning progressions 🧠 Equips teachers to analyze and respond to student thinking ▪️ Builds formative assessment into everyday practice ▪️ Supports schoolwide coherence, not just classroom-level changes Harvard article: https://lnkd.in/gfVRmb5R #mathed #WeLeadTexas #formativeassessment #ESSA #MTSS

9

2 Comments

So proud to see this innovative work from Jeremy Riel, UIC TRAILblazer Lab and the University of Illinois Chicago #CollegeofEducation! 🎉 #SkillTree is such an exciting example of how AI can support student learning and teacher feedback—using learner analytics to truly empower students. As a University of Illinois at Chicago - Graduate College alum, I’m incredibly proud and can’t wait to see #SkillTree in action and the impact it makes in classrooms!

6

1 Comment

Drum roll, please: Announcing EVAL Challenge finalists ... When we launched EVAL in September and the EVAL EdTech Challenge in October, we were eager to see how the field would respond. We hoped to learn whether there was shared interest in research partnerships focused on independent evaluation and understanding what works in real classrooms. The response exceeded anything we anticipated. We received 32 applications from remarkable organizations, ranging from early-stage startups to established companies serving learners across K–12 domains. In our conversations, we met thoughtful innovators deeply committed to improving learning and addressing long-standing opportunity gaps. This work has been made possible by a broad and growing community. Faculty across Boston University, including colleagues at Boston University Wheelock College of Education & Human Development and Hariri Institute for Computing, Boston University, played a central role in reviewing applications and will continue to support the next phases of research and partnership development. Our ecosystem includes educators, researchers, funders, accelerators, districts, and community partners, all committed to evidence-based, responsible AI innovation. We look forward to building many more partnerships as this ecosystem continues to grow. In this first Challenge cycle, EVAL will partner with one tool for a full randomized controlled trial. At the same time, our broader goal is to build shared research infrastructure that helps the field understand whether tools are effective, for whom they work, and under what conditions, while taking real-world implementation seriously. The work I am most passionate about is what comes next. Designing research that is rigorous yet practical, innovative yet cumulative, and directly useful for those making decisions about teaching, learning, and technology. I am deeply grateful to everyone who has engaged with EVAL so far. The real work is just beginning, and I am excited to keep building this together. We are always glad to connect with others thinking about these questions.

19