{"resourceId":"longbeach-lb911-commscoach-qa-city-20260223","versions":[{"version":"external-42ea534e13e3023b33bec309df7943594ea2e079f3cfca84510a83c71b075862","resource":{"id":"longbeach-lb911-commscoach-qa-city-20260223","title":"Long Beach city page: CommsCoach AI to score 100% of 911 calls for supervisor coaching","organization":"City of Long Beach Department of Disaster Preparedness and Emergency Communications (republishing ABC7)","sector":"Municipal emergency communications / public reporting","geography":"United States (Long Beach, California)","publishedAt":"2026-02-23","publicationDate":"2026-02-23","eventDate":"2026-02-22","sourceName":"Long Beach.gov Disaster Preparedness news","sourceLabel":"City of Long Beach news page on LB9-1-1 CommsCoach AI quality assurance","sourceUrl":"https://www.longbeach.gov/disasterpreparedness/news/lb9-1-1-leverages-ai-for-quality-assurance/","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["accessibility-workforce","governance-procurement","data-security","operating-model"],"finding":"A February 23, 2026 City of Long Beach news page (republishing ABC7, Feb 22) states LB9-1-1 will use GovWorx CommsCoach AI to review emergency calls and score dispatcher handling. The city reports ~600,000 911 calls/year; national standards target ≥2% QA (~12,000 calls) previously done manually. Director Reginald Harrison says the AI tool enables review of 100% of calls, with supervisors reviewing flagged calls for coaching. Contract cited at ~$68,000/year through February 2028; vendor claims ~250 other 911 centers use the tech. Harrison states AI will not replace dispatchers, take calls, or replace human judgment, and that AI listening stays within the city's secure data system after privacy/HIPAA vetting. Program described as in pilot phase.","sledRelevance":"Municipal primary URL (not previously archived; Hoodline secondary was) documenting 911 AI QA procurement, HITL coaching, and privacy claims. Why now: pairs with St. John's Aug 2026 AQA metrics as a second independent jurisdiction on full-coverage QA AI after recent non-emergency answering editions.","evidence":"Opened longbeach.gov HTML via curl. Release date 2026-02-23; 600,000 calls; 2%/12,000 standard; 100% review claim; CommsCoach/GovWorx; $68,000/yr to Feb 2028; HITL coaching and on-premises-data claim from page.","architectureImplications":"Plan AI QA as a scoring sidecar on call recordings with supervisor flag queues; keep scoring out of live dispatch decisioning unless separately validated.","governanceImplications":"Publish how scores enter evaluations, privacy report-card commitments, and pilot exit criteria before citywide performance use.","securityPrivacyImplications":"City claims AI listening stays in city secure systems under HIPAA-aligned vetting—verify residency, subprocessors, and deletion of unmapped sensitive fields in the contract.","caveats":"City/press announcement of a pilot—not measured post-deployment outcomes; 100% review claim is capability framing; Hoodline/privacy package already partially covered related facts.","streamIds":["emergency-services"],"roles":{"sales":"Position Long Beach's CommsCoach pilot—100% call scoring with supervisor coaching and an explicit no-replacement pledge—as a governance-aware 911 QA entry point. Discovery should cover current 2% sampling burden, privacy/HIPAA review status, and whether scores will touch formal evaluations. Bounded engagement: privacy-and-SOP readiness assessment plus pilot scorecard design. Unsupported claims include proven quality uplift, job reduction, or that city data-residency claims equal a completed independent audit.","engineering":"Fit is AI scoring of recorded 911 audio for QA flags into supervisor workflows. Architecture should keep inference/data in approved city or GovCloud boundaries per the city's stated model, with audit logs of scores and human overrides. Prerequisites: recording access, score dimensions (tone, protocol, judgment), and privacy controls. Constraints: pilot-stage; press metrics only. Proof of value needs flagged-call precision/recall versus supervisor gold and coaching throughput—not 100% coverage alone.","delivery":"Stand up pilot governance, supervisor coaching playbooks, and ongoing privacy monitoring for CommsCoach-flagged 911 calls. Dependencies include Disaster Preparedness/EOC leadership, city data security, and GovWorx. Ownership: ECC director for program; supervisors for coaching decisions. Acceptance: documented HITL on flags, data-residency attestation, and pilot KPIs before eval linkage. Risks include labor pushback and silent expansion from coaching into discipline."},"roleDetails":{"sales":{"customerProblem":"Large PSAPs cannot meet meaningful QA sampling without AI assist or more supervisors.","stakeholders":["ECC director","City privacy/security","Supervisors","Dispatcher workforce"],"discoveryQuestions":["Pilot vs production status?","Will scores enter evals?","Where does audio process?","What flag rate is tolerable?"],"valueHypothesis":"AI scoring may enable near-full QA coverage if supervisors still coach flags and privacy controls hold.","potentialEngagement":"Pilot governance and KPI design workshop before citywide eval use.","unsupportedClaims":["Proven quality improvement","AI replaces dispatchers","HIPAA claim equals completed audit"]},"engineering":{"fit":"Recording-based AI QA scoring with supervisor flag review.","architectureIntegration":"911 recorder → CommsCoach scores → supervisor flag queue → coaching/eval systems.","prerequisites":["Recording integration","Score rubric","City-approved hosting","Access logging"],"constraints":["Pilot phase","Press-only metrics","Sensitive PHI/PII on calls"],"security":"Enforce city data-system residency claims, HIPAA controls, and least-privilege access to scores/audio.","proofOfValue":["Supervisor agreement on flags","Coaching cycle time","QA coverage vs 2% baseline"]},"delivery":{"work":"Configure pilot, train coaches, and monitor privacy/labor impacts.","dependencies":["City security team","GovWorx","ECC supervisors"],"ownership":"Disaster Preparedness/ECC director owns program; supervisors own coaching.","skillsAdoption":"Teach supervisors to use flags as coaching prompts, not automatic discipline.","governanceCheckpoints":["Privacy review gate","Pilot exit criteria","Labor consultation"],"acceptanceCriteria":["HITL on flagged calls","No live-call replacement","Documented data residency"],"risks":["Eval mission creep","Privacy incident","Dispatcher distrust"]}},"retrievedAt":"2026-09-28T22:06:15-05:00","enrichedAt":"2026-09-28T22:06:15-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Reframes supervisor time from sampling toward coaching flagged calls; requires clear messaging that AI scores support development, not silent replacement.","procurementImplications":"Budget modest SaaS (~$68k/yr cited) plus privacy/security review time; require measurable pilot KPIs before expanding eval use.","operatingModelImplications":"Supervisors remain final coaches/evaluators; AI only prioritizes which of 100% scored calls get human attention.","sourceVerification":{"responseId":"curl-longbeach-commscoach-20260928T220615","openedUrl":"https://www.longbeach.gov/disasterpreparedness/news/lb9-1-1-leverages-ai-for-quality-assurance/","referenceExcerpt":"With this AI tool, we're actually able to review 100% of the calls,","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}