{"resourceId":"centralia-staff-ai-ordinance-20260922","versions":[{"version":"external-31531c1619f18620da56f9f14dc8b65085ab65920af395679cbbb7be70f2c099","resource":{"id":"centralia-staff-ai-ordinance-20260922","title":"Centralia unanimously adopts staff generative-AI ordinance requiring verification","organization":"City of Centralia","sector":"Small-city internal AI governance and workforce policy","geography":"Centralia, Missouri, United States","publishedAt":"September 22, 2026 KBIA public-radio report on Monday ordinance vote","publicationDate":"2026-09-22","eventDate":"2026-09-21","sourceName":"KBIA","sourceLabel":"Public-radio report on Centralia Board of Aldermen AI ordinance","sourceUrl":"https://www.kbia.org/kbia-news/2026-09-22/an-extra-layer-of-protection-centralia-adopts-ai-policy-for-city-staff","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["governance-procurement","data-security","accessibility-workforce","operating-model"],"finding":"On Sept. 22, 2026 KBIA reported Centralia’s Board of Aldermen unanimously passed an ordinance Monday regulating city staff use of generative AI. Staff must verify accuracy of any generated content; supervisors decide how far each department may use AI; unauthorized use can bring discipline or termination; AI-generated videos, images, and audio for city business require written supervisor permission; and protocols aim to stop confidential information from being entered into AI tools. City Administrator Tara Strain said AI has helped with paperwork and Facebook posts but will never replace police or manual labor. The piece cites a Tyler Technologies survey that only about a third of public-sector leaders have a defined AI policy, and quotes academic experts on legitimacy and scale-of-error risks.","sledRelevance":"Same-day small-city AI acceptable-use adoption—useful control checklist beside larger-city inventories and chatbot launches.","evidence":"KBIA Sept 22: unanimous ordinance; verification mandate; supervisor discretion; discipline; generative media permission; confidential-data protocols; Strain quotes; Tyler one-third citation.","architectureImplications":"Interpretation: Prioritize allowlisting and DLP-style confidential-prompt bans before broad Copilot rollouts.","governanceImplications":"Interpretation: Pair ordinance text with training and sampled verification evidence—passage alone is not control maturity.","securityPrivacyImplications":"Interpretation: Confidential-input bans and media-approval gates are the operational core of the reported ordinance.","caveats":"Article summarizes ordinance; full legal text not attached; Tyler survey is secondary.","streamIds":["local-government"],"roles":{"sales":"Small cities experimenting with generative AI need enforceable guardrails before tools spread informally. Centralia’s ordinance—verification of AI outputs, supervisor-scoped uses, confidential-data limits, and discipline paths—supports a policy-and-training engagement, not a product pitch that AI replaces police or manual labor (explicitly rejected by the city administrator). Bound work to acceptable-use policy drafting, tool allowlisting, and supervisor playbooks. Do not claim the Tyler one-third statistic is Centralia’s own survey.","engineering":"Fit is governance of staff generative-AI use across departments—not a resident-facing chatbot build. Architecture implications are allowlisted tools where feasible, logging or supervision hooks, and hard blocks on confidential prompts into public models. Prerequisites: ordinance or policy text, supervisor role definitions, and training. Constraints: small-city capacity; article lacks full legal text attachment. Security: stop confidential data entry into consumer AI and control generative media with written approvals.","delivery":"Adopt the ordinance into department SOPs, train staff on verification and confidential-data bans, and define supervisor approval workflows for AI media. Ownership: City Administrator with department supervisors. Skills: managers who can spot hallucinated content before public release. Governance checkpoints: 30-day compliance sample and exception log. Acceptance: staff can show verification steps; confidential-data incidents investigated. Risks: shadow AI continuing; uneven department rules; chilling useful low-risk use."},"roleDetails":{"sales":{"customerProblem":"Staff already use generative AI for paperwork and public posts while hallucination and confidential-data risks lack clear local rules.","stakeholders":["Board of aldermen","City administrator","Department supervisors","Police/public safety","Communications"],"discoveryQuestions":["Which tools are authorized per department?","How is verification evidenced?","What counts as confidential input?","How are generative media approvals logged?"],"valueHypothesis":"A lightweight ordinance plus supervisor discretion can reduce uncontrolled shadow AI if training and enforcement are real—without proving productivity gains.","potentialEngagement":"Policy drafting, allowlist, and supervisor training package for a small city—no automation platform sale required.","unsupportedClaims":["AI replaces police/manual labor","Guaranteed productivity savings","Tyler survey equals Centralia measurement","Ordinance eliminates hallucination risk"]},"engineering":{"fit":"Acceptable-use controls for staff generative AI with verification and confidential-data prohibitions.","architectureIntegration":"Supervisor-authorized tools → staff draft → mandatory human verification → optional written approval gate for AI media → publish/retain records.","prerequisites":["Written policy/ordinance","Tool allowlist","Training materials","Incident/discipline process"],"constraints":["Small-city IT capacity","Article is secondary to full legal text","Department-by-department variability","Consumer tools hard to monitor fully"],"security":"Prohibit confidential inputs to public models; require verification before external use; gate synthetic media.","proofOfValue":["Training completion","Exception/incident log trend","Sample of verified outputs","Supervisor approval records for AI media"]},"delivery":{"work":"Translate ordinance into SOPs, train all staff, and stand up supervisor approval and audit sampling.","dependencies":["Legal/ordinance text","HR/discipline procedures","IT allowlisting","Communications norms"],"ownership":"City Administrator owns policy; supervisors own day-to-day authorization.","skillsAdoption":"Teach staff to verify facts and never paste confidential data into public AI.","governanceCheckpoints":["Effective-date briefing","30-day compliance sample","Quarterly policy refresh","Incident review"],"acceptanceCriteria":["Staff acknowledge policy","Verification expectation documented","Confidential-data ban communicated","AI media permission path exists"],"risks":["Uneven enforcement","Shadow AI persists","Over-blocking low-risk uses","Public messaging errors if verification skipped"]}},"retrievedAt":"2026-09-23T03:11:30Z","enrichedAt":"2026-09-23T03:15:54Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Interpretation: Policy should not freeze helpful drafting aids if verification and non-digital service capacity (e.g., police staffing) remain explicit.","procurementImplications":"Interpretation: Small cities may buy policy/training help before enterprise AI platforms.","operatingModelImplications":"Interpretation: Supervisor-scoped rules require consistent management capacity across departments.","sourceVerification":{"openedUrl":"https://www.kbia.org/kbia-news/2026-09-22/an-extra-layer-of-protection-centralia-adopts-ai-policy-for-city-staff","referenceExcerpt":"when city staff use AI, they must be able to verify the accuracy of any generated content.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}