Artificial Intelligence Governance of Predictive Risk-Scoring Models Under Unmeasured Community Consent Withdrawal in Health Insurance Enrollment: A Healthcare Political Legitimacy Evaluation

Predictive risk-scoring models used in health insurance enrollment allocate access to care on the basis of inferred future cost, yet the communities whose data train those models have no mechanism for withdrawing consent that the model can register. the present analysis evaluates the political legitimacy of that arrangement. It argues that the legitimacy deficit is not a data-protection compliance problem but a problem of unmeasured consent withdrawal: the withdrawal exists, is politically expressed, and is structurally invisible to the scoring pipeline. It then characterises the operating characteristics of a before-after-control-impact design that a regulator could use to detect the effect of consent withdrawal on enrollment outcomes, using a declared variance structure with an intraclass correlation of 0.64, between-cluster SD of 6.0 and residual SD of 4.5. Power was estimated across 2000 replications per design cell for cluster counts of 8, 16, 24 and 32 and effects of 3, 5 and 8 percentage points; 54 clusters are required for 80% power at a 5 percentage point effect. The contribution is a political-theoretic account of why consent withdrawal must be measured at community level before risk-scoring governance can claim legitimacy, together with a design that makes that measurement feasible.