Personalized for Whom? Putting AI’s Biggest Promise to Its Hardest Test

Personalized learning is the central promise of AI in education, but a decade in, a pointed question is emerging: personalized for whom, and to what level of success? AI-driven personalization has historically served students best when they perform near grade level and generate the kind of data these models were trained on. It faces far greater strain for students in MTSS Tier II and III interventions, multilingual learners and students receiving special education services, supports most AI personalization tools were never designed to deliver. This poster presentation examines whether the evidence holds up. Explore where AI personalization tools show the most evidence of impact and where that evidence thins as student needs grow more complex. Learn to distinguish tools genuinely engineered for the full range of learners from those offering surface-level differentiation, with a practical framework for asking better procurement questions. Hear what district leaders are discovering as pilots expand, and leave with a clearer picture of what responsible AI product development requires in training data, model design and implementation support.