Cooling with consent: Toward equitable and readiness-based AI-guided heat adaptation

Urban heat is intensifying worldwide, yet many AI-driven planning tools focus primarily on spatial optimisation while overlooking the social, behavioural, and institutional dynamics that eventually determine whether climate adaptation measures succeed or fail. This paper argues that technical optimisation alone is an insufficient condition for effective urban heat adaptation. For interventions to be durable, legitimate, and widely adopted, they must align with societal behavioural readiness, institutional feasibility, participatory mechanisms, and strong equity safeguards. We therefore advocate shifting from conventional risk-based approaches toward a readiness-based framework that prioritises where interventions are most socially and institutionally viable. Unlike traditional concepts of resilience or adaptive capacity, the proposed framework operationalises readiness through measurable behavioural, participatory, and institutional indicators, while explicitly integrating physically derived outputs from urban climate modelling, such as thermal comfort indices and temperature projections, as technical foundations for the assessment process. By embedding these dimensions within AI-enabled tools, the framework translates contextual readiness into actionable intelligence while safeguarding the interests of vulnerable populations. Drawing on behavioural science, institutional theory, and environmental justice, we reframe the core planning issue from ‘where the heat is most severe’ to ‘where intervention is most feasible, equitable, and inclusive’. The paper outlines an interdisciplinary research agenda and presents a staged implementation pathway, including data requirements and governance considerations to support real-world application. As a conceptual contribution, empirical validation through future pilot studies remains a necessary next step.