{"resourceId":"ladotd-i12-ai-queue-detection-no-retention-2026","versions":[{"version":"external-af77ad1f358c69f9a067a3d30ca6f5d62d22fec51833c1163e1c14f06b44f72a","resource":{"id":"ladotd-i12-ai-queue-detection-no-retention-2026","title":"Louisiana DOTD I-12 AI queue-detection pilot confirmed not to record video, stills or alert logs","organization":"Louisiana Department of Transportation and Development (DOTD)","sector":"State DOT interstate traffic management / queue warning","geography":"I-12 westbound near I-12/I-10 merge, Denham Springs / Baton Rouge area, Louisiana, United States","publishedAt":"2026-08-03 (DOTD written confirmation via Corzo); deployment reported June 29-30, 2026","publicationDate":"2026-08-03","eventDate":"2026-06-29","sourceName":"Chris Corzo Injury Attorneys (reporting DOTD written responses); corroborated by KPEL/WBRZ","sourceLabel":"Attorney publication of DOTD Q&A on I-12 AI queue-detection records retention","sourceUrl":"https://www.callcorzo.com/i-12-warning-system-camera-footage/","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["data-security","governance-procurement","infrastructure","operating-model"],"finding":"Louisiana DOTD's first camera-based AI queue-detection system on westbound I-12 (near the I-12/I-10 merge after College Drive) measures vehicle speeds and traffic frequency to warn drivers of forming backups via flashing prepare-to-stop signs, without facial recognition or license-plate logging per June 2026 local reporting. In an Aug 3, 2026 publication, Chris Corzo Injury Attorneys released DOTD's written answers stating the pilot cameras do not record video or still images and do not retain queue alerts, traffic measurements, warning-light activations, TMC logs or operator notes for later crash requests. Why-now: addresses the 2026-11-05 edition's named gap on U.S. state DOT AI operations and privacy design.","sledRelevance":"Rare documented U.S. state DOT posture that roadside AI traffic cameras can be scoped to ephemeral queue detection without building a surveillance archive—useful for DOTs answering privacy objections while piloting AI incident/queue tools.","evidence":"Corzo Aug 3, 2026 Q&A table quoting DOTD (representative name omitted; firm says it retains correspondence). KPEL June 30 and WBRZ June 29 describe first-in-state camera AI queue detection and no facial recognition/LPR.","architectureImplications":"Design queue AI as real-time speed/volume analytics driving DMS activation without persistent media storage; separate any enforcement cameras into different systems with explicit retention rules.","governanceImplications":"Publish retention matrices for AI roadway cameras before pilots; answer FOIA/discovery questions with written scope limits.","securityPrivacyImplications":"Non-retention reduces privacy and discovery risk but also limits after-action learning from video; agencies must still secure live camera views in the TMC.","caveats":"Attorney-published correspondence rather than a DOTD press PDF; no published detection-time or crash-reduction metrics; June deployment news is older than the last edition date.","streamIds":["transportation"],"roles":{"sales":"Sales framing: Louisiana's I-12 AI queue pilot shows a state DOT deploying camera analytics for earlier backup warnings while confirming—in writing—no video, still or alert retention. For U.S. DOT buyers facing privacy pushback, the bounded value is scoping AI to ephemeral speed/volume detection and DMS activation without building an enforcement archive. This does not prove crash reductions or that every DOTD camera statewide shares the same rules.","engineering":"Engineering framing: implement queue detection as streaming speed/occupancy inference that triggers warning signs, with storage disabled for media and alert histories unless a separate logged system is approved. Keep LPR/face models out of the path. Prerequisites include calibrated detection zones, DMS integration and TMC operator workflows. Confirm retention settings in procurement and acceptance tests so defaults cannot silently start recording.","delivery":"Delivery framing: pilot one merge/queue hotspot, publish a retention matrix, train TMC staff on live-view-only use, and measure warning lead time versus prior radar/sensor baselines before expansion. Dependencies include DOT legal for records policy and 511/DMS operations. Acceptance requires verified non-retention configuration and public FAQ. Risks include public confusion with enforcement cameras and inability to retrieve footage after crashes."},"roleDetails":{"sales":{"customerProblem":"State DOTs want AI queue/incident detection for safety but face surveillance and records-request concerns about roadside cameras.","stakeholders":["DOT secretary/TMC leadership","State privacy/legal counsel","Motorists and civil-liberties groups","Legislators","Maintenance/ITS contractors"],"discoveryQuestions":["Do queue cameras record or only stream?","What alerts and measurements are retained?","Is LPR/face detection enabled?","How will crash litigants be told no footage exists?"],"valueHypothesis":"An ephemeral AI queue-warning design can deliver earlier DMS activation while remaining politically and legally defensible if non-retention is verified and published.","potentialEngagement":"Policy-and-architecture review to draft a camera retention matrix and pilot acceptance tests for non-enforcement AI traffic detection.","unsupportedClaims":["Proven crash or congestion reductions on I-12","That all Louisiana cameras follow this pilot's rules","Transfer of attorney-reported Q&A without DOTD primary PDF"]},"engineering":{"fit":"Interstate TMC queue warning and wrong-way/slowdown detection programs that must avoid becoming surveillance systems.","architectureIntegration":"Camera to real-time analytics to DMS/TMC alert; no media archive; configuration-as-code for retention flags.","prerequisites":["Detection-zone calibration","DMS control integration","Written retention standard","Operator live-view procedures"],"constraints":["Limited performance metrics published","Attorney secondary publication of DOTD answers","Pilot geography only"],"security":"Protect live TMC camera access; ensure analytics hosts cannot be reconfigured to record without change control.","proofOfValue":["Warning lead time vs prior sensors","Rear-end crash proxy metrics","Configuration audit showing storage off","Public FAQ accuracy check"]},"delivery":{"work":"Deploy pilot zone, lock non-retention settings, publish FAQ, evaluate warning effectiveness, decide expansion.","dependencies":["DOT legal/records","TMC operations","ITS vendor","Communications"],"ownership":"DOTD owns policy and ops; vendor owns analytics under DOT change control.","skillsAdoption":"Train operators that no replay exists; train public affairs on privacy messaging.","governanceCheckpoints":["Retention matrix sign-off","Pre-go-live config audit","30/90-day performance review","Expansion decision"],"acceptanceCriteria":["DOT-confirmed non-retention","No LPR/face features enabled","DMS activation tested","Public documentation live"],"risks":["Discovery disputes if any recording is found later","Safety critics wanting retained video","Scope creep into enforcement uses"]}},"retrievedAt":"2026-11-06T22:12:15-06:00","enrichedAt":"2026-11-06T22:17:30-06:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Earlier queue warnings can protect all drivers including commercial and aging road users; TMC staff must understand live-only workflows without replay.","procurementImplications":"Specify non-retention and no-biometric analytics in ITS contracts with audit rights on storage configuration.","operatingModelImplications":"Separate safety queue-warning AI from enforcement camera programs with distinct policies and hardware.","updateExplanation":"June deployment reporting predates last run; included because Aug 3 DOTD written non-retention confirmation newly documents U.S. state DOT privacy controls for roadway AI—explicit gap from the 2026-11-05 edition.","peerReview":"unknown","sourceVerification":{"openedUrl":"https://www.callcorzo.com/i-12-warning-system-camera-footage/","referenceExcerpt":"do not record video or capture still images.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}