[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81594-en":3,"doc-seo-81594-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},81594,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",7,"Healthcare","Empowering 9-1-1 Calltaking Training with Generative AI: Experiences and Lessons Learned","Emergency call-takers handle over 240 million calls annually yet face a training crisis driven by persistent staffing shortages and high per-trainee instruction time, while manual quality assurance covers fewer than 10% of calls and delays feedback. Partnering with Metro Nashville Department of Emergency Communications, a GenAI-powered training system was deployed in real operations for six months, scaling to 190 users across 1,120 sessions. Analysis of deployment logs and artifacts yields four lessons with concrete design and governance practices for responsible, human-centric AI training in safety-critical public environments.","Empowering 9-1-1 Calltaking Training with Generative AI: Experiences and Lessons Learned  \nZirong Chen  \nCollege of Connected Computing Vanderbilt University Nashville, TN [zirong.chen@vanderbilt.edu](zirong.chen@vanderbilt.edu)  \nMeiyi Ma  \nCollege of Connected Computing Vanderbilt University Nashville, TN [meiyi.ma@vanderbilt.edu](meiyi.ma@vanderbilt.edu)  \narXiv :2602 . 1324 1v 3 [ cs .CY] 9 Jul 2026  \nAbstract—Emergency call-takers form the first operational linkin public safety response, handling over 240 million calls annually while facing a sustained training crisis: staffing shortages exceed 25% in many centers, and preparing a single new hire can require up to 720 hours of one-on-one instruction that removes experienced personnel from active duty. Traditional training approaches struggle to scale under these constraints, limiting both coverage and feedback timeliness. In partnership with Metro Nashville Department of Emergency Communications (MNDEC), we designed, developed, and deployed a GenAI-powered calltaking training system under real-world constraints1 Over six months, deployment scaled from initial pilot to 190 operational users across 1,120 training sessions, exposing systematic challenges around system delivery, rigor, resilience, and human factors that remain largely invisible in controlled or purely simulated evaluations. By analyzing deployment logs capturing 98,429 user interactions, organizational processes, and stakeholder engagement patterns, we distill four key lessons, each coupled with concrete design and governance practices. These lessons provide grounded guidance for researchers and practitioners seeking to deliver AI-driven training systems in safety-critical public sector environments where practical constraints fundamentally shape human-centric design.  \nIndex Terms—Emergency response training, Responsible AI deployment, Human-AI collaboration, High-stakes systems  \nI. INTRODUCTION  \nEmergency call-takers handle over 240 million calls annually in the United States [3], serving as critical first responders who coordinate multi-agency emergency response under extreme time pressure [15] . These operators make lifeor-death decisions navigating complex protocol trees while managing distressed callers, yet face an escalating training crisis. Centers nationwide report staffing shortages exceeding 25 percent, with some jurisdictions approaching 40 percent deficits [22] . Training each operator requires up to 720 hours of intensive instruction, removing experienced personnel from active duty for months [2] . Manual quality assurance covers fewer than 10 percent of calls, creating significant delays  \n1This work is a collaboration between the Vanderbilt Research Team and the Metro Nashville Government, including Metro Nashville Information Technology Services and the Metro Nashville Department of Emergency Communications. We sincerely thank our partners for their exceptional efforts in providing domain expertise and professional evaluation to support a responsible design, development, and deployment process. We are especially grateful to the IT support, training, quality assurance, and management teams at the Metro Nashville Department of Emergency Communications.  \nbetween performance and feedback [16], [22], [37] . GenAI systems promise scalable solutions through automated scenario generation, realistic caller simulation, and consistent performance assessment [6], [21] . However, deploying GenAI within government emergency centers exposes challenges absent from controlled research settings. Prior work identifies adoption barriers in public sector AI including infrastructure limitations, organizational resistance, and skills gaps [9] . Responsible AI frameworks offer governance principles emphasizing transparency, fairness, and accountability [23] . Yet existing literature lacks longitudinal empirical evidence of sustained deployment navigating the embedded constraints characteristic of safety-critical","cbCaibwluo4Az03B","https://ap.wps.com/l/cbCaibwluo4Az03B","pdf",1603521,3,1,8,"English","en",105,"# Introduction\n## Operational challenge and training crisis\n## GenAI opportunities and responsible governance gaps\n## Longitudinal deployment approach\n## System functions and scaling results","[{\"question\":\"What training crisis motivates this work for 9-1-1 call-takers?\",\"answer\":\"Emergency centers face sustained staffing shortages exceeding 25% in many places, and preparing a new hire can require up to 720 hours of intensive one-on-one training. Limited manual quality assurance (under 10% of calls) creates delayed performance feedback.\"},{\"question\":\"What does the GenAI-powered calltaking training system automate?\",\"answer\":\"It automates generating realistic emergency scenarios across 57 incident types and assessing trainee performance against 1,651 protocol requirements. The system uses telephony and audio interfaces to role-play callers and deliver just-in-time debriefing after sessions.\"},{\"question\":\"How was the system deployed and evaluated over time?\",\"answer\":\"After an initial pilot across 18 instrumented workstations, deployment scaled over six months to 190 operational users completing 1,120 training sessions. The evaluation analyzes deployment logs of 98,429 user interaction events and additional system events and audio recordings, alongside development artifacts and stakeholder co-design.\"}]",1784174586,20,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"empowering-9-1-1-calltaking-training-with-generative-ai-experiences-and-lessons-learned","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/healthcare/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/empowering-9-1-1-calltaking-training-with-generative-ai-experiences-and-lessons-learned/81594/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-21","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What training crisis motivates this work for 9-1-1 call-takers?","Question",{"text":75,"@type":76},"Emergency centers face sustained staffing shortages exceeding 25% in many places, and preparing a new hire can require up to 720 hours of intensive one-on-one training. Limited manual quality assurance (under 10% of calls) creates delayed performance feedback.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What does the GenAI-powered calltaking training system automate?",{"text":80,"@type":76},"It automates generating realistic emergency scenarios across 57 incident types and assessing trainee performance against 1,651 protocol requirements. The system uses telephony and audio interfaces to role-play callers and deliver just-in-time debriefing after sessions.",{"name":82,"@type":73,"acceptedAnswer":83},"How was the system deployed and evaluated over time?",{"text":84,"@type":76},"After an initial pilot across 18 instrumented workstations, deployment scaled over six months to 190 operational users completing 1,120 training sessions. The evaluation analyzes deployment logs of 98,429 user interaction events and additional system events and audio recordings, alongside development artifacts and stakeholder co-design.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,118,122,126,129,133],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":116,"slug":117},40,"healthcare",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":119,"show_sort_weight":120,"slug":121},"Research & Report",30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":29,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":29,"doc_module":4,"doc_module_name":46,"category_name":127,"show_sort_weight":29,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":46,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":46,"category_name":135,"show_sort_weight":106,"slug":136},19,"General","general"]