Most U.S. states have established AI governance frameworks with central technology teams setting initial policies and oversight, but responsibility often shifts to individual agencies once AI systems are deployed. This decentralized operational accountability requires agencies to monitor performance, manage risks, and address issues, while central offices provide standards and oversight, though variations exist across states like Maryland, California, and Pennsylvania. A key challenge remains in defining clear ownership and accountability for AI tools post-deployment, especially as AI increasingly appears embedded in existing software, complicating governance and requiring ongoing scrutiny during procurement and use.
https://www.govtech.com/spotlight/50-states-50-different-ways-who-owns-ai-once-its-deployed