A clinic chatbot retains chat logs in ways that raise surveillance and privacy concerns for vulnerable residents. What ethical issue is highlighted?

Public-health chat assistants handle sensitive conversations. Retaining clinic chatbot logs in ways that raise surveillance and privacy concerns for vulnerable residents highlights privacy and surveillance ethics: prolonged or poorly governed retention can chill care-seeking, expose health-related disclosures, and feel like monitoring rather than help. Algorithmic bias and privacy concerns in public-health AI assistants are a different issue and do not erase the privacy and surveillance concerns raised by how clinic chat logs are retained, since the problem here is not biased outputs even though privacy is mentioned. Algorithmic bias and privacy concerns in public-health AI assistants fail because the ethical problem here is prolonged or poorly governed log retention that feels like monitoring, not biased or unfair model outputs. Informed consent and privacy limits in public-health AI assistants fail because the question concerns how chat logs are retained for surveillance-like monitoring, not whether patients understood or voluntarily agreed to AI-assisted care. Surveillance and transparency gaps in public-health AI assistants fail because the highlighted issue is surveillance-like log retention and privacy, not a lack of transparency or explainability for decisions, even though surveillance is mentioned. Exam caveat: some retention may be required for safety, quality, or legal holds; the ethical issue is disproportionate or unclear retention that functions as surveillance, not a ban on all logging. Operational check: minimize fields retained; set short default retention with access controls; publish a plain-language notice of what is stored and why; exclude open-data publication of chat content; review logs only under defined clinical or compliance roles.