The municipal AI operating model includes a model-risk and ethics review board for sensitive resident services. What design choice is illustrated?

Sensitive resident services—housing eligibility, benefits adjudication, enforcement triage, child and family supports—demand structured model-risk and ethics review before scale, not after harm. Embedding an ethics and model-risk review board into the municipal AI operating model is the design choice illustrated: review is a standing capability with charter, intake criteria, escalation paths, and decision rights, not an optional favor. That works because civic AI failures are often socio-technical: biased features, unclear human override, opaque notices, or unsafe automation of consequential decisions. A board institutionalizes challenge, documentation, and cross-disciplinary scrutiny (legal, privacy, equity, operations) inside the operating model rather than leaving ethics to after-the-fact crisis response.

Skipping ethics review because sensitive services are assumed low risk fails by concept: sensitivity and consequentiality are precisely why risk is elevated, not why review can be waived. Reviewing ethics only after a public scandal forces a reaction fails because reactive boards are remedial theater; they do not prevent the first wave of resident harm or restore trust cheaply. Leaving ethics to individual staff judgment with no board or process fails because unaided individual judgment is inconsistent, conflicted, and unauditable at city scale.

Exam caveat: candidates may think ethics review slows innovation; on CAIPM-style items, responsible operating models treat review as enabling safe scale for sensitive services. Keep the municipal framing—resident services and civic accountability are the stake. Operational check: verify the board’s charter covers sensitive uses, sample recent intakes for documented risk findings and mitigations, and confirm production releases for those services show completed review rather than silent exceptions.