This brief from the Center on Reinventing Public Education at Arizona State University examines how a subset of “Early Adopter” school districts and charter management organizations are integrating artificial intelligence (AI) into their systems. The authors argue that districts are making consequential decisions about AI adoption, governance, and instructional integration largely on their own. Drawing on survey and interview data from 45 districts across 20 states, the study finds that although Early Adopters have become more sophisticated in their approaches, most are using AI to improve existing instructional models rather than fundamentally rethink teaching and learning. The findings also suggest that AI is exposing the limitations of traditional change management and that the broader support infrastructure around districts often reinforces existing approaches rather than encouraging new models. This brief may benefit teachers and teacher preparation faculty by providing insight into the system-level decisions and conditions shaping how AI is introduced and used in schools.