The New York Post reports that Florida-based public safety technology company Aurelian has built an AI call-taker called Ava that answers police non-emergency phone lines, takes reports, and routes complaints. According to The New York Post, Aurelian CEO Max Keenan told The California Post that "roughly 70%" of calls entering a typical emergency communications center are non-emergencies, including barking-dog complaints, parking disputes, lost property reports, and requests for city services. The New York Post also cites a March report finding that the LAPD answered just 57.43% of 911 calls within California's 15-second standard in 2024, below the state benchmark. City officials told The New York Post that roughly 100 operators must be on duty to meet minimum staffing requirements. The story frames the AI pitch as aimed at alleviating non-emergency load on strained dispatch centers.
What happened
The New York Post reports that Florida public safety technology company Aurelian has developed an AI call-taker named Ava that answers police non-emergency phone lines, takes reports, and routes complaints. According to The New York Post, Aurelian CEO Max Keenan told The California Post that "roughly 70%" of calls entering a typical emergency communications center are non-emergencies. The New York Post also cites a March report that found the LAPD answered just 57.43% of 911 calls within California's 15-second standard in 2024. The New York Post reports city officials estimate about 100 operators must be on duty across a 24-hour period to meet minimum staffing requirements.
Technical details
Editorial analysis - technical context: The article lists the kinds of low-acuity call types Ava handles, barking dogs, parking disputes, noise complaints, lost property, abandoned vehicles, and suspicious-activity reports, which are typically structured, repeatable interactions. For practitioners, these call types map well to automation techniques that combine automatic speech recognition, intent classification, slot-filling workflows, and deterministic routing to municipal service channels or online forms. Industry experience suggests success on such tasks depends on high transcription accuracy in noisy environments, robust intent taxonomies, and clear handoff triggers to human dispatchers when ambiguity or safety signals appear.
Context and significance
Emergency dispatch centers nationwide face staffing shortages and high volumes of non-emergency traffic, which public reporting connects to slower emergency response times. The article frames Aurelian's pitch as an operational relief tool for strained centers rather than a replacement for emergency dispatch. For technology teams evaluating similar deployments, the key technical and governance concerns include false negatives/positives on emergency detection, auditability of automated decisions, and integration with existing CAD (computer-aided dispatch) and records systems.
What to watch
Observers should track independent performance evaluations, published accuracy and safety metrics for emergency-detection logic, and any municipal pilot results or procurement records. Also watch for vendor disclosure of escalation thresholds, human-in-loop workflows, and data retention or privacy policies as municipalities consider adopting automated call-takers.
Scoring Rationale #
Notable operational deployment with direct implications for public-safety workflows and practitioners building voice/triage systems. The story is relevant for teams designing ASR, intent detection, and human-in-loop escalation, but it does not present a technical breakthrough.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.