In addition to the cost of developing and maintaining enterprise-scale applications, employee turnover, software erosion, and technical debt make ongoing maintenance increasingly challenging and potentially fraught with error as applications age. TxDOTCONNECT (the Texas Department of Transportation’s custom-built system for managing the delivery of transportation programs, projects, and right of way) is a multi-modular application that operates across more than 25 environments and presents key opportunities for AI-augmented streamlined maintenance and operations. In this session, we’ll outline how a standard AI framework can be designed to enhance the speed of AI application development while ensuring the availability of a governed and secure model repository. In addition, we’ll provide a little bit of inspiration for how incremental investment in AI augmentation—such as by starting with narrow use cases—can deliver massive value and reduce exorbitant operational costs, all while keeping AI-generated risk minimal.