Systemic readiness of AI solutions in health: structuring a multicriteria model to assess the adequacy between solution and context

The adoption of Artificial Intelligence (AI) solutions in healthcare depends not only on good algorithmic performance, but also on the ability to function safely and usefully in the context in which it will be used. Traditional maturity assessment instruments capture only part of this problem: the TRL scale assesses the maturity of the technology, while models such as HAIRA assess the capability of the organization. This work proposes a multicriteria model to assess the systemic readiness of AI solutions in health, understood as the adequacy between a solution and its application context. Thus, the pair (solution, context) is adopted as the unit of evaluation. In each criterion, what one party requires is compared with what the other is able to offer, identifying gaps that may limit adoption. The criteria are organized into five dimensions, associated with fundamental objectives, and critical criteria can exercise veto, making the model partially non-compensatory.