{"resourceId":"nbc26-brown-county-dispatch-ava-ai-handles-60pct-non-emergency-calls-20261028","versions":[{"version":"external-c2a0010109059c02df7ec9b44e691166c461f689069e811472f3f04f8ff50b5a","resource":{"id":"nbc26-brown-county-dispatch-ava-ai-handles-60pct-non-emergency-calls-20261028","title":"Brown County, Wis., dispatch puts an AI assistant on its non-emergency line, handling about 60% of those calls while humans keep all 911 calls","organization":"Brown County Public Safety Communications (Wisconsin)","sector":"911 emergency communications and dispatch","geography":"Brown County (Green Bay area), Wisconsin, United States","publishedAt":"2026-10-28","publicationDate":"2026-10-28","eventDate":null,"sourceName":"NBC 26 (WGBA, Green Bay)","sourceLabel":"NBC26: Brown County dispatch employs AI to answer non-emergency calls","sourceUrl":"https://www.nbc26.com/news/local-news/brown-county-dispatch-employs-ai-to-answer-non-emergency-calls","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["operating-model","accessibility-workforce","knowledge-work"],"finding":"NBC 26 reports that Brown County Public Safety Communications, which receives about 160,000 calls a year on its non-emergency number, went live this month with A.V.A. (Automated Virtual Assistant) to answer non-critical calls such as records requests and parking permits. The center spent six months adapting the tool with help from La Crosse, Waukesha and Dane County centers that already use it. Operations manager Jessica Skalecki says A.V.A. follows dispatcher operating guidelines, detects frustration or background concerns and transfers callers with a transcript, and that dispatchers monitor all A.V.A. calls in real time. The center says A.V.A. handles about 60% of non-emergency calls on average, will not replace staff, and will never answer 911 calls.","sledRelevance":"Another Wisconsin county joins a regional cluster of centers sharing the same non-emergency AI assistant, showing peer-to-peer adoption among U.S. PSAPs and a firm local line that 911 calls stay human.","evidence":"Local TV report quoting the operations manager and director. The 60% figure is the operator's own average from the first weeks of operation; no baseline, error rate, abandoned-call rate or staff-time saving is given, and 'overwhelmingly good response' is anecdotal.","architectureImplications":"AI answers a separate non-emergency line, with real-time dispatcher monitoring and warm transfer that passes a transcript to the human; 911 trunks are untouched.","governanceImplications":"The center publicly commits that AI will not answer 911 calls and that staff are not being replaced; agencies can turn these into written policy and review metrics.","securityPrivacyImplications":"Not discussed; non-emergency calls include records requests and personal details, so transcript retention and vendor access need terms.","caveats":"Early operator claims in the first month of live use; vendor not named in the article; 60% is an average without a defined denominator period or quality measure. Wisconsin context similar to earlier archived county deployments, so the novelty is a new jurisdiction rather than new evidence.","streamIds":["emergency-services"],"roles":{"sales":"For sales, the framing point is that Brown County adopted non-emergency AI by borrowing from three nearby Wisconsin centers and now reports about 60% of those calls handled. Directors under call-volume pressure respond to peer references and a clear promise that 911 stays human. A bounded engagement is a measured pilot design for a non-emergency line. The report cannot support claims of staff savings, accuracy or caller satisfaction beyond anecdote.","engineering":"For pre-sales engineering, the design is AI on a separate non-emergency line with live dispatcher monitoring, frustration detection and transcript handoff. Prerequisites are telephony routing, transcript delivery into the dispatcher workflow and records-request integrations. Constraints include keeping 911 trunks separate and state records rules. Proof of value should compare containment, transfer accuracy and abandoned calls with a pre-launch baseline.","delivery":"For delivery, the work is configuring call flows to local operating guidelines, setting transfer triggers and running live monitoring. The six-month local build suggests real effort even with peer help. Ownership sits with the communications center operations manager. Acceptance should rest on measured containment, correct transfers and dispatcher workload. Risks include callers stuck in the bot, missed urgent calls on the non-emergency line and drift in the 60% figure."},"roleDetails":{"sales":{"customerProblem":"Non-emergency calls (about 160,000 a year here) pull dispatchers away from 911.","stakeholders":["Communications center director","Operations manager","County board","Dispatcher staff and union","Neighbouring centers already using the tool"],"discoveryQuestions":["How many non-emergency calls do you receive and what types?","Which neighbouring centers could share configuration or lessons?","How would you measure containment and transfer quality?"],"valueHypothesis":"AI handling of routine non-emergency calls might free dispatcher time for 911 if transfers work reliably (interpretation; the source gives no time-saved data).","potentialEngagement":"A 90-day measured pilot on one non-emergency line with baseline and transfer-quality review.","unsupportedClaims":["Staff hours saved","Error or misrouting rates","Caller satisfaction beyond anecdote","Suitability for 911 calls"]},"engineering":{"fit":"Fits PSAPs with a distinct non-emergency number and high routine-call volume.","architectureIntegration":"AI answers the non-emergency line, dispatchers monitor live, and transfers carry a transcript to the human.","prerequisites":["Separate non-emergency telephony routing","Transcript handoff into dispatcher workstation","Local operating guidelines encoded in call flows"],"constraints":["911 trunks must remain human-answered","State public-records and retention rules","Vendor not named in source"],"security":"Limit transcript retention and vendor access; audit who reviews recorded calls.","proofOfValue":["Containment rate against defined denominator","Share of transfers judged correct on review","Abandoned-call and repeat-call rates versus baseline"]},"delivery":{"work":"Configure call flows, transfer triggers and live monitoring; train dispatchers on handoffs.","dependencies":["Telephony configuration","Records and permit system access","Peer-center configuration sharing"],"ownership":"Communications center operations manager under the director.","skillsAdoption":"Dispatchers learn to monitor AI calls and pick up transfers using transcripts.","governanceCheckpoints":["Written rule that 911 stays human","Weekly review of transferred and failed calls","Public notice of AI use"],"acceptanceCriteria":["Proposed: containment rate reported monthly with method","Proposed: no urgent call left unescalated in audit sample"],"risks":["Urgent calls on non-emergency line missed","Caller frustration with bot","Overstated containment figures"]}},"retrievedAt":"2026-10-28T21:59:46-05:00","enrichedAt":"2026-10-28T22:04:35-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Framed as freeing dispatchers for 911 rather than cutting jobs; callers who prefer a person can ask for a dispatcher, which matters for older or limited-English callers.","procurementImplications":"Shows neighbouring-county references and shared configuration help a center adopt the same tool; buyers should still require their own performance data.","operatingModelImplications":"Splits non-emergency intake to AI with live human oversight while 911 call-taking stays fully human.","sourceVerification":{"responseId":"curl-brown-2026-10-28","openedUrl":"https://www.nbc26.com/news/local-news/brown-county-dispatch-employs-ai-to-answer-non-emergency-calls","referenceExcerpt":"A.V.A. is currently handling about 60% of the center's non-emergency calls on average.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}