{"resourceId":"seattle-q2-2026-ai-usage-report","versions":[{"version":"external-c191f3dd24ef8098d3391d71ceb3d9fc3225cc99a68b50393d272fb37c33dca0","resource":{"id":"seattle-q2-2026-ai-usage-report","title":"Seattle Q2 AI report pairs favorable CivCheck proof-of-value with unspent permitting funds","organization":"City of Seattle Information Technology Department","sector":"Municipal IT and construction permitting","geography":"Seattle, Washington, United States","publishedAt":"Quarterly report dated July 1, 2026","publicationDate":"2026-07-01","eventDate":"2026-07-01","sourceName":"Seattle City Clerk filing / Seattle IT Artificial Intelligence Usage Report 2026 Q2","sourceLabel":"Official city quarterly AI usage and pilot evaluation report","sourceUrl":"https://clerk.seattle.gov/~cfpics/cf_323759_q2.pdf","evidenceClass":"government-evaluation","outcomeClass":"mixed","topics":["governance-procurement","operating-model","accessibility-workforce","knowledge-work"],"finding":"SDCI completed a CivCheck proof-of-value finding automated checks accurate and capable of reducing intake screening times by up to 46 days, yet $750,000 appropriated for an AI permitting tool remained unspent while several pilots were still being evaluated.","sledRelevance":"New-to-archive official status report showing that favorable component evaluation does not automatically trigger procurement spend or production deployment in a major U.S. city.","evidence":"Executive summary and Lessons Learned state CivCheck PoV completion focused on middle housing dwellings, with automated checks described as accurate, comprehensive, and capable of reducing intake screening times by up to 46 days. Financial section states SDCI has not moved forward on implementing an AI permitting tool yet while evaluating several pilots, and that no amount of the $750,000 Construction and Inspections Fund appropriation had been spent to date. Broader inventory notes 21 approved AI software in active use and 18 pilots/PoCs. The report does not republish the full Innovation Hub methodology tables in the inspected Lessons Learned excerpt.","architectureImplications":"Interpretation: Keep proof-of-value environments separable from production permitting systems until funding, supportability and scale criteria are met.","governanceImplications":"Interpretation: Use council-facing quarterly reporting to make the gap between pilot findings and purchase decisions visible to oversight bodies.","securityPrivacyImplications":"Interpretation: Continue applying the city’s Responsible AI review and accuracy/hallucination audit criteria before any permitting production path; the Q2 narrative does not replace a security assessment.","caveats":"Distinct from the already-archived Innovation Hub CivCheck study URL; this filing adds budget and deployment status. 'Up to 46 days' is a city-reported upper bound without the full comparator table in the excerpted Lessons Learned section. PDF text extraction may miss figures. Copilot usage stats are separate from permitting outcomes.","streamIds":["local-government"],"roles":{"sales":"Interpretation: City IT and construction departments may have completed favorable pilots while appropriations remain unused. Include budget owners, SDCI leaders, labor partners and the AI review team in discovery. Ask what supportability and funding gates remain after PoV. A useful engagement clarifies evaluation criteria and operating costs rather than pushing immediate license spend. The value hypothesis is a documented go/no-go with residual risk, not automatic savings equal to the 'up to 46 days' figure. Do not claim Seattle has deployed CivCheck citywide. Confirm whether multiple pilots are still competing before proposing a single vendor path.","engineering":"Interpretation: Treat PoV accuracy claims as inputs to a production design review covering coverage of correction types, integration with intake systems and reviewer workload. Prerequisites include the post-pilot evaluation packet, funding authority and Responsible AI sign-off. Constraints include parallel pilots and unspent appropriations signaling unresolved fit. Security should re-test data flows for permit applications before go-live. Proof of value for production must restate clocks, coverage and supportability—not only component accuracy.","delivery":"Interpretation: Keep a city service owner accountable for the decision memo linking PoV results to spend or deferral. Dependencies include council reporting cadence, labor engagement and competing pilot results. Acceptance criteria should require an explicit production recommendation, spent-versus-appropriated tracking and user-experience thresholds from the city’s PoV framework. Train staff on interim manual processes while evaluation continues. Risks include political pressure to spend unused funds without supportability evidence, or stalling despite favorable metrics without documenting why."},"roleDetails":{"sales":{"customerProblem":"Permitting backlog pressure coexists with unused AI appropriations and incomplete production decisions.","stakeholders":["Construction & Inspections leadership","City IT / Responsible AI program","Budget and council staff","Labor partners","Pilot vendors under evaluation"],"discoveryQuestions":["What remains unresolved after the CivCheck PoV?","Which supportability and funding gates block spend?","Are multiple pilots still competing?","How will quarterly public reporting treat a deferral?"],"valueHypothesis":"Structured go/no-go analysis can convert pilot findings into an accountable procurement or deferral decision without equating PoV metrics to live savings.","potentialEngagement":"Decision-support engagement that maps PoV evidence to production prerequisites and residual risks.","unsupportedClaims":["Seattle has production-deployed CivCheck","Guaranteed 46-day savings after purchase","Unused funds imply vendor failure rather than deliberate evaluation"]},"engineering":{"fit":"Post-PoV production readiness assessment for permitting pre-screening tooling.","architectureIntegration":"Reconnect PoV findings to intake systems, identity, logging and reviewer workflows only after go-live criteria are met.","prerequisites":["Complete post-pilot evaluation packet","Funding authority","Responsible AI recommendation","Defined production success metrics"],"constraints":["Parallel pilots","Unspent appropriation","Council reporting expectations"],"security":"Re-validate application-data handling, vendor access and audit logging before any production path.","proofOfValue":["Restated clocks and coverage for production scope","Supportability/scalability sign-off","User-experience threshold checks","Spend-versus-benefit tracking plan"]},"delivery":{"work":"Produce an explicit deploy, defer or re-compete decision with public-facing rationale.","dependencies":["Quarterly AI usage reporting","Labor and department coordination","Competing pilot evidence"],"ownership":"SDCI business owner with Seattle IT AI Review Team recommendation path.","skillsAdoption":"Maintain reviewer capacity and interim procedures while evaluation continues.","governanceCheckpoints":["After PoV packet complete","Before any appropriation spend","Before production applicant exposure"],"acceptanceCriteria":["Written production recommendation","Appropriation spend status explained","No silent scope expansion beyond evaluated use"],"risks":["Spending to exhaust funds","Indefinite stall without documented rationale","Confusing PoV upper bounds with guaranteed outcomes"]}},"retrievedAt":"2026-09-15T14:13:55Z","enrichedAt":"2026-09-15T14:15:10Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Interpretation: Continue workforce training and labor engagement noted in the report before any permitting production change that alters staff or applicant workflows.","procurementImplications":"Interpretation: Keep appropriations and vendor selection gated on post-pilot supportability evidence; unspent funds can be a control signal rather than delay alone.","operatingModelImplications":"Interpretation: Separate innovation-fund and permitting-tool budget lines from automatic deployment authority after a favorable PoV narrative.","updateExplanation":"New-to-archive July 1, 2026 clerk-filed quarterly report; adds deployment and appropriation status beyond the earlier Innovation Hub CivCheck study already in the archive.","sourceVerification":{"openedUrl":"https://clerk.seattle.gov/~cfpics/cf_323759_q2.pdf","referenceExcerpt":"the pilot found that CivCheck’s automated checks were accurate, comprehensive, and capable of reducing intake screening times by up to 46 days.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}