Priority Quality Attributes for Engineering AI-enabled Systems

Deploying successful software-reliant systems that address their mission goals and user needs within cost, resource, and expected quality constraints require design trade-offs. These trade-offs dictate how systems are structured and how they behave and consequently can effectively be evolved and sustained. Software engineering practices address this challenge by centering system design and evolution around delivering key quality attributes, such as security, privacy, data centricity, sustainability, and explainability. These concerns are more urgent requirements for software-reliant systems that also include AI components due to the uncertainty introduced by data elements. Moreover, systems employed by the public sector exhibit unique design time and runtime challenges due to the regulatory nature of the domains. We assert that the quality attributes of security, privacy, data centricity, sustainability, and explainability pose new challenges to AI engineering and will drive the success of AI-enabled systems in the public sector. In this position paper, we enumerate with examples from healthcare domain concerns related to these requirements to mitigate barriers to architecting and fielding AI-enabled systems in the public sector.