Residents cannot learn why a benefits chatbot denied a suggestion, and staff raise transparency concerns. What ethical issue is illustrated?

Residents and staff deserve reasons, not opaque denials. When people cannot learn why a benefits chatbot denied a suggestion, the ethical issue is lack of transparency and explainability in civic AI. Benefits guidance affects access to support; unexplained refusals prevent residents from correcting information, appealing, or understanding eligibility, and they prevent staff from auditing the bot’s logic. Lack of algorithmic fairness and equity as an ethical concern in civic AI is not the issue illustrated, because the scenario describes a failure to provide any case-specific reason for the denial, not evidence of biased decisions or inequitable outcomes for different groups. Lack of strict accountability and oversight as an ethical concern in civic AI is a separate concern; the scenario shows no failure of human responsibility or review structures, only a general failure of transparency and explainability when residents and staff cannot learn why a suggestion was denied. Lack of data privacy and confidentiality as an ethical concern in civic AI is about protecting personal information from improper disclosure, not the complete absence of a reason; the scenario shows residents and staff cannot learn why the denial occurred at all, not that private data was exposed. Exam caveat: explainability must be meaningful and accurate, not a fabricated rationale. A fluent wrong reason is still an ethical and operational failure; explanations should map to policy rules and logged inputs. Operational check: require the chatbot to return a resident-facing reason category tied to policy; log underlying decision features for staff review; escalate unexplained or contested denials to a human benefits worker before the case is closed.