Beyond the Balance Sheet: Tangible and Intangible Return on Investment in Radiology Artificial Intelligence Adoption

Artificial intelligence (AI) products for radiology are being cleared by the US Food and Drug Administration (FDA) at a pace that has outrun the field's ability to evaluate them financially. Hospital leadership, radiology directors, and other members on AI governance committees are routinely asked to justify high upfront and recurring costs using a return-on-investment (ROI) vocabulary built for capital equipment with a direct revenue line - a vocabulary that fits only a minority of the products now on the market. This paper argues that AI product ROI, when properly assessed, is not limited to departmental cash flows and must include both tangible value (billable revenue, throughput, avoided cost) and intangible value (diagnostic confidence, workforce retention, referring-physician trust, institutional reputation, and strategic optionality). This paper also reasons that the balance between tangible and intangible value is not constant across AI products or timelines. Instead, this balance shifts somewhat predictably with the underlying technology architecture and its associated use case, from narrow triage tools with fast, mostly tangible payoffs to platform and language-model products whose value is more diffuse and harder to separate from the surrounding workflow. Lastly, this paper proposes a governance framework - anchored by a proxy-measurement table that operationalizes both tangible and intangible value across eleven institutional domains - to help radiology leadership evaluate tangible and intangible ROI together at the point of purchase and track how each is expected to move across the product's operating life, ensuring deployment and renewal decisions are judged against realistic expectations for each individual AI technology for its specific use case.