{"resourceId":"longbeach-commscoach-qa-privacy-2026","versions":[{"version":"external-9765422cbbb46bb7bb9680555943bdc70bf382fdf09b0033bb50e10ab3ba2a67","resource":{"id":"longbeach-commscoach-qa-privacy-2026","title":"Long Beach pilots CommsCoach AI to score 911 calls for coaching with privacy guardrails","organization":"City of Long Beach Disaster Preparedness and Emergency Communications / GovWorx","sector":"911 / Emergency Communications Center quality assurance (post-call)","geography":"Long Beach, California, United States","publishedAt":"2026-04-29","publicationDate":"2026-04-29","eventDate":"2026-04-29","sourceName":"Hoodline","sourceLabel":"Independent local reporting summarizing city privacy materials and press coverage","sourceUrl":"https://hoodline.com/2026/04/ai-judge-joins-long-beach-911-center-city-swears-it-is-just-for-coaching/","evidenceClass":"independent-reporting","outcomeClass":"mixed","topics":["governance-procurement","data-security","accessibility-workforce","operating-model"],"finding":"Hoodline (Apr 29, 2026) reports Long Beach is piloting GovWorx CommsCoach to transcribe and score recorded 911 calls for supervisor coaching (~600,000 annual calls; historic ~2% manual QA; ~$68,000/year through Feb 2028), with privacy materials citing AWS GovCloud processing, 24-hour deletion of non-mapped audio, CJIS/HIPAA framing, and human-final performance decisions—while officials say the AI will not take live calls; some dispatcher skepticism about automated scoring is noted.","sledRelevance":"Why now: not previously archived; newly relevant after 2026-09-17 emergency-path and non-emergency PSAP AI catalogue entries—documents a post-call QA control pattern with explicit retention/cloud guardrails as a lower life-safety-risk alternative to live triage AI.","evidence":"Apr 29, 2026 publish; 600k calls/year; ~2% historic review; CommsCoach; $68k/year through Feb 2028; post-call only; AWS GovCloud; non-mapped audio deleted within 24 hours; CJIS/HIPAA claims; supervisor-final evaluations; union/dispatcher skepticism cited.","architectureImplications":"Offline/post-call transcription and rubric scoring with supervisor review queue—not real-time dispatcher prompts or caller-facing agents.","governanceImplications":"Publish scoring validity, false-flag rates, and labor consultation outcomes; keep humans as final arbiters of employee performance.","securityPrivacyImplications":"GovCloud processing, short deletion of non-mapped audio, access auditing, and CJIS/HIPAA claims are positive controls that still need independent verification.","caveats":"Local aggregation; no published accuracy study of scores; April date predates last run but URL was new to archive; policing/fire share ECC—focus is 911 QA controls.","streamIds":["emergency-services"],"roles":{"sales":"Position CommsCoach-style QA as a governance-friendly entry point for ECC AI—expanding review coverage without putting AI on live emergency audio paths. Discovery should verify GovCloud/retention claims in the city’s privacy package and ask for scoring false-positive rates. Offer a bounded QA-coverage pilot with labor consultation. Unsupported claims include proven coaching outcome gains, replacement of supervisors, or endorsement of live-call AI from this story.","engineering":"Fit is post-call QA scoring with human review, not live triage. Architecture needs secure recording ingest, GovCloud processing, deletion timers, and supervisor workflow. Prerequisites include CJIS pathways and union notice. Constraints: scoring validity unpublished. Proof of value is increased reviewed-call share with acceptable false-flag rates and no live-call scope creep.","delivery":"Pilot on recorded calls only; calibrate flags with supervisors; keep performance decisions human. Ownership sits with ECC QA leadership and city privacy officers. SkillsAdoption trains supervisors to override bad scores. Governance checkpoints include privacy package attestation and labor check-ins. Risks include treating scores as automatic discipline and expanding into live monitoring without new PIAs."},"roleDetails":{"sales":{"customerProblem":"ECCs review too few 911 calls for QA while facing pressure to adopt AI without live-path life-safety risk.","stakeholders":["ECC directors","QA supervisors","City privacy/IT","Labor unions"],"discoveryQuestions":["Is scope post-call only?","What is the false-flag rate?","Where is audio processed/deleted?","Who finalizes employee scores?"],"valueHypothesis":"If AI QA increases reviewed-call share under human-final decisions and verified retention controls, agencies may fund a bounded pilot—without claiming live triage benefits.","potentialEngagement":"Bounded post-call QA coverage pilot with privacy and labor package.","unsupportedClaims":["Proven coaching outcomes","Supervisor replacement","Live-call AI safety","Guaranteed staffing relief"]},"engineering":{"fit":"Post-call transcription/scoring for supervisor coaching.","architectureIntegration":"Recording store → CommsCoach → flag queue → supervisor UI; no live caller path.","prerequisites":["CJIS-compliant path","Deletion automation","Access audit logs","Rubric calibration set"],"constraints":["No published score validity","Pilot calibration ongoing","Labor sensitivity"],"security":"Enforce GovCloud, retention timers, least-privilege access, and vendor training limits.","proofOfValue":["Reviewed-call percentage","False-flag sample rate","Supervisor override rate","No live-scope drift"]},"delivery":{"work":"Enable post-call pilot; calibrate rubrics; train overrides; attest privacy controls.","dependencies":["Recording system","GovWorx contract","Privacy office","Union notice"],"ownership":"QA supervisor owns coaching use; privacy officer owns retention compliance.","skillsAdoption":"Teach supervisors to treat scores as hypotheses, not verdicts.","governanceCheckpoints":["Privacy attestation","30-day calibration review","Labor feedback session"],"acceptanceCriteria":["Post-call-only confirmation","Documented deletion SLA","Human-final performance rule"],"risks":["Automated discipline","Live-monitoring creep","Unvalidated rubrics"]}},"retrievedAt":"2026-09-19T03:11:37Z","enrichedAt":"2026-09-19T03:16:37Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"May increase coaching bandwidth for understaffed QA; dispatcher skepticism highlights workforce trust risk if scores feel punitive.","procurementImplications":"~$68k/year through 2028 should be tied to coverage/false-flag KPIs and privacy SLA verification, not seat counts alone.","operatingModelImplications":"Shifts QA from sparse manual listening toward AI-filtered supervisor review without changing live call-taking.","sourceVerification":{"openedUrl":"https://hoodline.com/2026/04/ai-judge-joins-long-beach-911-center-city-swears-it-is-just-for-coaching/","referenceExcerpt":"non-mapped audio is deleted within 24 hours","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}