{"resourceId":"gunpo-ai-legislative-support-rag-20260928","versions":[{"version":"external-5b820f740d056ea3282d1a0e7fa93f088d4227c892fcb28efd6e8469c68c23a9","resource":{"id":"gunpo-ai-legislative-support-rag-20260928","title":"Gunpo City Council kicks off RAG-LM legislative-support research with human-verification standard through November","organization":"Gunpo City Council, Gyeonggi Province, South Korea (via Aju Press)","sector":"Local legislative AI support / ordinance and budget review","geography":"Gunpo, Gyeonggi Province, South Korea (transferable methods with jurisdictional limits)","publishedAt":"September 28, 2026","publicationDate":"2026-09-28","eventDate":"2026-09-23","sourceName":"Aju Press","sourceLabel":"Independent reporting / council research kickoff","sourceUrl":"https://www.ajupress.com/view/20260928112870567","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["knowledge-work","governance-procurement","operating-model","data-security"],"finding":"Reported Sept. 28, 2026: Gunpo City Council’s AI Innovation Research Group held a Sept. 23 kickoff for a study through November titled “Establishing an AI Legislative Support System Based on RAG-LM and Legislative Innovation Measures.” The work examines AI assistance for searching legislative materials and reviewing ordinances, budgets, and policies, emphasizing RAG-LLM over pre-provided council materials with verification of related evidence. Scope includes measuring effects on search time, efficiency, accuracy, and justification quality, plus risks (privacy/security, data quality, over-reliance). The council will review guidelines based on the principle that AI-generated results should not be used directly for decision-making without human verification; representative Park Sang-hyun stressed verifying real assistance value while keeping human final judgment.","sledRelevance":"Fresh same-day international local example of RAG-grounded legislative AI with explicit human-verification—transferable design pattern for U.S. councils studying ordinance/budget copilots, with clear transferability limits.","evidence":"Opened Aju Press article dated September 28, 2026 describing the Sept. 23 kickoff and November study window. No measured accuracy or time-savings results yet—research phase.","architectureImplications":"RAG over council-controlled corpora rather than open-web generation; evidence verification and human final judgment as design requirements before decision use.","governanceImplications":"Research-first path with safe-usage standards before production; human verification principle stated up front for ordinances and budgets.","securityPrivacyImplications":"Study scope includes data privacy/security and data-quality risks; implies controlled corpora and limits on over-reliance.","caveats":"Kickoff/research announcement, not a deployed production system; no published metrics yet; South Korean municipal law and council process differ from U.S. home-rule contexts—transfer methods, not statutes.","streamIds":["local-government"],"roles":{"sales":"Council clerks and legislative staff want AI that finds ordinance and budget materials without silently deciding. Gunpo’s Sept. 23 research kickoff (reported Sept. 28) studies a RAG-LM legislative support system through November and states AI-generated results must not drive decisions without human verification. Sell a bounded discovery or pilot for retrieval quality and human-final workflows—not autonomous ordinance drafting or guaranteed time savings before evaluation finishes.","engineering":"Fit is retrieval-augmented legislative research over a controlled council corpus. Prerequisites: curated materials, citation or evidence display, privacy review, and a hard human-approval gate before any recommendation enters the official record. Constraints: research-phase only; cross-border legal differences; accuracy metrics not yet published. Prove value with search-time and citation-accuracy samples against librarian baselines, plus zero unverified AI text in adopted findings.","delivery":"Delivery is a time-boxed research program: build RAG scenarios for ordinance and budget review, measure efficiency and accuracy, draft responsible-use standards, and keep human final judgment on every consequential output. Dependencies: council members, legislative staff, IT and security, and legal advisors. Ownership: research-group leads with council oversight. Risks: demo theater; over-reliance; importing the Korean process unchanged into a U.S. charter city."},"roleDetails":{"sales":{"customerProblem":"Legislative staff lose time searching ordinance and budget materials and risk unverified AI drafts entering deliberations.","stakeholders":["Council leadership","Legislative research staff","IT/security","Legal counsel","Clerk"],"discoveryQuestions":["What corpus will ground answers?","How are citations shown to members?","What is banned from AI-final decisions?","How will accuracy be scored by November?"],"valueHypothesis":"A RAG legislative assistant with mandatory human verification can cut search time while keeping ordinance and budget judgment with humans.","potentialEngagement":"Eight-week legislative RAG discovery: corpus map, citation UX, accuracy sample plan, and human-verification SOP.","unsupportedClaims":["Guaranteed ordinance throughput gains before study results","That RAG alone prevents hallucination","Direct transfer of Korean council procedure to U.S. cities"]},"engineering":{"fit":"Strong for grounded search and drafting aids; weak for autonomous voting recommendations or legal advice.","architectureIntegration":"RAG-LLM over pre-provided council materials; evidence/citation layer; no direct write into official decisions without human approval.","prerequisites":["Controlled corpus","Access control","Evaluation harness","Human verification workflow"],"constraints":["Study ends November—no production metrics yet","Privacy of non-public materials","Cross-jurisdiction transfer limits"],"security":"Keep corpus and prompts inside approved systems; assess data quality and over-reliance; log who accepted AI-assisted text.","proofOfValue":["Search-time vs baseline","Citation precision/recall sample","Rate of human edits before use","Zero unverified AI text in adopted documents in sample"]},"delivery":{"work":"Run kickoff scenarios, build evaluation sets for ordinance/budget tasks, draft usage standards, report November findings.","dependencies":["Council research group","IT","Legal","Records/archives"],"ownership":"AI Innovation Research Group owns study; council owns any later adoption.","skillsAdoption":"Members and staff practice verifying sources rather than pasting AI prose into motions.","governanceCheckpoints":["Corpus sensitivity review","Mid-study accuracy check","Human-verification standard draft","Go/no-go for production"],"acceptanceCriteria":["Documented scenarios","Published evaluation method","Written human-verification rule","November findings memo"],"risks":["Premature production launch","Over-reliance on fluent answers","Ignoring transferability limits"]}},"retrievedAt":"2026-09-28T22:20:57-05:00","enrichedAt":"2026-09-28T22:27:03-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"If deployed, legislative AI must not replace member deliberation capacity; training on verification is part of the stated research scope.","procurementImplications":"Any later build should require RAG grounding on council materials, evaluation deliverables, and contractual human-in-the-loop—not open-ended chatbot seats.","operatingModelImplications":"Research-then-standardize model for local councils: measure assistance value before authorizing AI in ordinance/budget workflows.","sourceVerification":{"openedUrl":"https://www.ajupress.com/view/20260928112870567","referenceExcerpt":"AI-generated results should not be used directly for decision-making without human verification.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}