Supporting emergency dispatchers with AI reinforcements

Emergency contact centers are experiencing a quiet crisis.

This and other troubling findings were published in America’s 911 Workforce in Crisis survey report, released last year by the International Academies of Emergency Dispatch (IAED) and the National Association of State 9-1-1 Administrators (NASNA). This joint report based on survey results from emergency contact centers around the country, highlighted troubling trends: from 2019 until 2022, the average vacancy rate in 911 centers across America hovered around 25%. Nearly every agency surveyed reported staffing turnover in 2022.

Staffing shortages affected every type of 911 center, rural and urban, large and small, across 48 states. The answers followed a clear pattern, with high stress, low wages, and long hours listed among the top concerns. A majority of contact centers responded that they were combating the issue in the short term with mandatory overtime (70%) or voluntary overtime (85%). One dispatch employee replied that in order to maintain emergency service availability in the face of high turnover, “we are unable to offer our employees the mandated breaks they deserve,” unfortunately compounding the issue by escalating burnout and increasing turnover.

“While these measures can be effective in the short term,” the report concludes, “they are not sustainable in the long run. Agencies simply must find long-term solutions to the staffing problem if they are to serve the public well.”

How AI can help

We’ve covered the ways that AI can be used to improve public sector experience for both citizens and employees, or even find and create new efficiencies. In addition, AI is providing hope to emergency call centers. Some predict that by integrating AI technology into emergency dispatch, 911 centers can relieve overworked staff, handle unexpected call surges, and provide faster service to citizens, all with one solution.

While it might seem futuristic, systems like this are already being used across the country, with municipalities in Colorado, Maryland, Missouri, Oregon, South Carolina, Texas and Virginia testing or implementing AI into their 911 centers. When someone calls the Charleston County's Consolidated Emergency Communication Center non-emergency line, for example, Amazon Connect will redirect the caller to appropriate resources – for example, routing a caller who is concerned about a wild animal directly to animal control. The system has reduced the volume of calls to the administrative line by 36% since implementation. Jefferson County in Colorado, which processes around a million calls per year, says that AI has processed nearly 40% of the emergency center’s administrative calls.

AI-enhanced but human driven

For all AI’s promise, many do have misgivings about the prospect of calling 911 and reaching a robot instead of a human.

Emergency contact centers will have to consider public reception. According to a study from the Pew Research Center, “A larger share of Americans say they are ‘more concerned than excited’ about the increased use of AI in everyday life than say they are ‘more excited than concerned’ about these prospects (37% vs. 18%). And nearly half (45%) say they are equally excited and concerned.” As one recent study on AI perception phrased it, concerns surrounding AI were driven by “human favoritism, rather than AI aversion.”

“If these systems are [designed] to take calls, rather than assisting call takers, it would remove a human empathy that is so often essential in crisis situations,” Daniela Gilbert, the director of the Vera Institute of Justice’s Redefining Public Safety Initiative, wrote in an email to GovTech. “Imagine being in a time of stress and great need and having to negotiate with a bot.”

AI solutions work best as complement or enhancement to human workers, not a replacement. For example, TTEC Digital’s AutoDirect Service uses AI to triage non-emergency calls by assisting human operators. A human operator answers every 911 call to assess if it’s an emergency. If the call does not require emergency dispatch, the operator offers the caller the help of the non-emergency bot. If the caller consents, the AutoDirect Service helps the caller find the information they need. This can include transferring the call to a different department, texting the caller a link to resources, providing documents, scheduling an appointment, collecting feedback, or providing information on events, emergency procedures, contact information, project status updates — whatever it has been programmed to do.

This entire scenario is AI-enhanced, but ultimately humans make every single decision. The operator must decide that the call can be resolved with AutoDirect Service, and the caller must consent. At any point during the call, the caller can opt to return to a live 911 agent, or the operator can return, escalating the call back to human hands.

“I cannot stress enough that it does not take away jobs, especially in the 911 industry. It's there to help them enhance their job,” said Jacob Saur, the emergency communications center administrator for Arlington County Public Safety Communications and Emergency Management.

Helping 911 evolve

America’s 911 Workforce in Crisis concludes that recruiting more applicants, retaining employees, and improving conditions will all be necessary to keep staffing in 911 centers level with demand. While AI-powered tools can’t solve every issue currently facing emergency contact centers, they can directly assist in several of the report’s key recommendations, including ensuring that the demands and expectations of emergency dispatch jobs are manageable and allowing for rising call volumes.

There’s a lot at stake. As emergency call centers continue to grapple with understaffing issues, some 911 calls may go unanswered or get stuck in lengthy queues. According to Andrew Dameron, Denver’s director of Emergency Communications: "If your residents lose faith that dialing 911 will enable them to get the help that they need, then you start to erode faith in the entire system.”