← Cases
United StatesPublic sector2025 – ongoing

Metropolitan Government of Nashville and Davidson County

Generative AI emergency-call training and automated debriefing

Nashville's AI call-taking trainer reached 190 users across 1,120 training sessions

WorkingCompany's own claim

Problem

Emergency call-taker training required experienced staff to spend time away from active duty. Trainees also needed timely feedback on their performance.

Approach

Vanderbilt researchers and Metro Nashville developed and deployed a generative AI training system. Models played callers through a telephone interface, and the system assessed responses against protocol requirements. Trainees received debriefing after each session.

Results

190operational users after 6 months, reported by the project researchersCompany's own claim[1]
1,120training sessions over the deployment period, reported by the project researchersCompany's own claim[1]

Takeaway

Deployment logs established sustained training use. Session counts did not establish improved handling of real emergency calls.

Categories

Business function
Human resources
Innovation type
Process innovation
Evidence
Company's own claim

Sources

Source summaries

[1]arXiv9 Jul 2026

The system's developers described a 6-month deployment with Nashville's emergency communications department. They reported 190 operational users and 1,120 training sessions.

Full details are in the original sources.

Added 1 Nov 2026 · Updated 1 Nov 2026
Open case page ↗