Artificial Intelligence Has Arrived at 911 and Is Changing How Emergencies Are Handled in the US
911 is one of those rare services where an actual human still picks up the phone when you call. No bot, no hold queue, no endless menu of options. But that is starting to change, and in a way that is worth paying close attention to.
Artificial Intelligence has arrived at emergency dispatch centers across the United States, and it did not come to replace dispatchers. It came to give real support to the people working on the front lines, helping them handle an overwhelming volume of calls that grows every year while staffing levels fail to keep up.
Think about it: roughly 240 million calls per year pour into these centers, and a huge chunk of them are not actual emergencies. The result? Overworked dispatchers, skeleton crews, and a turnover rate that would alarm any manager.
That is exactly where AI enters the picture, helping to filter non-emergency calls, provide real-time language translation, transcribe calls instantly, and even pinpoint, in a flood of hundreds of calls, the one person who truly needs urgent help.
Let us break down how this technology is transforming emergency dispatch in the US and what it could mean for the future of first-responder services. 🚨
911 Became a Victim of Its Own Success
For generations, 911 operators have been the backbone of America’s emergency response network. They take distress calls, dispatch and coordinate police, firefighters, SWAT teams, mental health counselors, and even animal control officers. In a sense, 911 ended up becoming a victim of its own success: the system is so reliable and so well-known that everyone knows they can call when they need help.
And call they do. That is where those 240 million annual calls we mentioned come in. But what a lot of people do not realize is that those same operators are also responsible for simultaneously handling non-emergency calls, which actually outnumber real emergencies. In New York, for example, there is the 311 line for that kind of demand. In Los Angeles, it is 877-ASK-LAPD.
These non-emergency calls, while not crises, are still important. People call to complain about parking issues, barking dogs, fireworks, minor fender benders, stolen bikes, missing cats, and abandoned cars. They also call to file police reports, request an extra patrol in the neighborhood, ask for help installing car seats, or swap out smoke detector batteries.
Aurelian, one of the companies offering AI assistance for dispatchers, found that 64% of all calls that reach dispatch centers are not real emergencies. On top of that, 70% of those non-emergency calls can be handled by AI without any human involvement. To give you an idea of the scale, in New York City alone the 311 line received more than 1.3 million calls just in March.
The Problem Nobody Could Solve Alone
Call volume has grown dramatically over the years, but staffing simply has not kept pace. Today, an experienced dispatcher might spend hours managing dozens of simultaneous calls, many involving extremely high-stress situations, which takes a serious toll on mental health and quality of life.
As Stephen Kennedy, the 911 coordinator in Sumter County, Florida, who handles about 80,000 calls a year, put it: when someone calls 911, that person does not want to be put on hold. But how do you guarantee that nowadays?
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
The most worrying detail is that a significant share of those non-emergency calls eats up precious time that could go toward real life-threatening situations. People calling to complain about noisy neighbors or report issues that other channels could handle end up competing for attention with someone having a heart attack or being the victim of a crime. It is not the callers’ fault, of course, but it is a structural problem that dispatch centers needed to tackle.
Professional burnout is common among dispatchers. A 2023 study by the International Academies of Emergency Dispatch found a 25% vacancy rate at dispatch centers across the country. According to Max Keenan, CEO and co-founder of Aurelian, 80% of new dispatchers quit within two years. That is a staggering turnover rate that shows just how demanding this profession really is.
Keenan used a pretty direct analogy to describe the situation: you basically train your 911 operators to be Navy SEALs, and they spend 70% of their time being mall cops. Constantly switching between routine calls and real emergencies is incredibly draining.
How Artificial Intelligence Is Working in Practice
The most advanced AI systems can listen to and interact directly with callers, in some cases taking over non-emergency calls while staying alert for any sign that the situation is actually a real crisis. Other systems provide automatic language translation to make sure anyone can get immediate help regardless of the language they speak.
It is important to note that in the vast majority of cases, AI is not directly answering 911 calls. It functions more as a support tool, relieving pressure on the trained dispatchers who frequently work in understaffed centers.
Aurelian, for example, already serves about 5 million Americans daily in places like Hamilton County in Tennessee, Kalamazoo in Michigan, and Grant County in Washington. The veteran Motorola Solutions, known for its police radio systems, also offers AI-assisted dispatch services. And Axon, the body camera manufacturer, provides AI-powered cameras that offer real-time translation services.
At Denver’s 911 dispatch center, professionals say they are exploring a variety of AI options, although they still have questions about cost and reliability, which is completely understandable when you are dealing with technology applied to life-or-death situations.
Intelligent Call Triage
The most straightforward AI application in these centers is automated call triage. Tools powered by advanced language models can identify within the first few seconds of a call whether the situation requires an immediate emergency response or can be routed to another service channel. This does not mean the call gets ignored. It means the call gets routed more efficiently, freeing dispatchers to focus where the risk is real and immediate.
In a demonstration for USA TODAY, the Aurelian system answered a call from a man reporting his neighbor’s barking dogs. The system interacted with him the same way a human operator would, gathering information about the address, how often the barking happened, and whether he wanted an officer sent to the scene.
Spotting Real Emergencies Hidden in the Noise
One of the most striking examples of how AI can make a difference happened during a power outage in an area served by Aurelian. About 600 people called the non-emergency line within a two-hour window to report the blackout. But buried in that avalanche of calls was one person asking for help because a critical medical device they depended on to survive had stopped working. The AI system recognized that this person needed immediate attention and connected them to a 911 operator.
That kind of ability to find a needle in a haystack is something that would be extremely difficult for an overloaded human to do consistently with hundreds of calls coming in at the same time.
Real-Time Automatic Translation
Another area where AI is making a huge difference is automatic translation. The United States is an extremely linguistically diverse country, and callers do not always speak English fluently enough to describe an emergency clearly.
Before AI, these situations relied on human translators who needed to be located and patched into the call. According to Kennedy in Sumter County, depending on the language, it could take five to ten minutes to get a human translator on the line. With AI, the person simply speaks in their own language and the system translates automatically.
As he rightly pointed out, imagine the anxiety and stress for both the person answering and the person calling while you are trying to let them know help is on the way but you cannot communicate.
Automatic Call Transcription and Summarization
In Florida, dispatchers in Sumter County have been happy with the Motorola AI system that automatically transcribes and summarizes every call that comes into 911. This allows other dispatchers to share that information in near real time with first responders heading to the scene of an accident or a shooting.
The system can also flag critical keywords like cardiac arrest or active shooter, letting other team members start mobilizing emergency resources even while the original operator is still gathering information from the caller. This ability to act in parallel can significantly cut down response times.
Keeping Humans at the Center of the Operation
At the National Emergency Number Association, or NENA, the team is closely monitoring how dispatch centers are using AI and how the public is responding to it. There is a legitimate concern, especially given the negative press around hastily deployed AI systems caught hallucinating facts or delivering incorrect information in unexplainable ways. Cost is also a relevant factor.
April Heinze, NENA’s vice president of operations and standards, emphasized that it remains essential for someone calling 911 to report an emergency to hear a human voice on the other end of the line, someone who conveys safety, competence, and empathy.
People want to talk to another person, as she put it simply.
Heinze explained that some dispatch centers are experimenting with AI to triage the massive volume of calls that 911 receives during significant public events. Think about a major car pileup on a busy highway or a wildfire sending up a huge column of smoke visible for miles. In those situations, 911 operators need to focus on dispatching emergency services to the scene, but they end up fielding hundreds of identical calls from well-meaning people who are simply reporting the same event.
With AI, a dispatch center could temporarily route all 911 calls from that specific area to an automated system. That system would ask callers whether they are just reporting that accident or fire, whether they have specific information that first responders need to know, or whether they are calling about a different emergency. This smart filtering frees human operators to manage the actual response to the ongoing event.
It is important to be clear that AI does not replace the human element in this context, and it probably never will completely. The ability to pick up on emotional nuances, make complex ethical decisions in split seconds, and offer basic psychological support to someone in a panic are skills that belong to humans. What the technology does is amplify those capabilities, cutting through the operational noise and allowing dispatchers to be more present, more attentive, and more effective in the moments that matter most. 🤝
Who Is Developing These Solutions
Beyond Aurelian, Motorola Solutions, and Axon, other companies specializing in public safety technology are also investing heavily in this transformation. Platforms that integrate AI directly into dispatch center workflows are becoming increasingly common, connecting data from devices like smartwatches and health apps straight to dispatchers. This enriches call context with information that simply did not exist at the time of the call before.
Smaller startups are also entering this market with solutions focused on specific niches, like serving hearing-impaired populations or integrating security camera feeds to help dispatchers see the scene while they are still on the phone. The innovation ecosystem around emergency dispatch services is growing fast.
It is worth noting that the development of these technologies does not happen in a vacuum. There are real concerns about privacy, algorithmic bias, and transparency in AI-assisted decision-making processes. Companies operating in this space need to demonstrate that their systems were trained on representative data and that when errors occur, they do not put vulnerable populations at even greater risk.
What Changes for the People on the Front Lines
For dispatchers, the arrival of Artificial Intelligence represents something many of them have already described as genuine relief. Not because the technology solves every problem, but because it lifts an operational burden that drained their energy without necessarily adding value to the handling of real emergencies.
When an automated system can filter non-emergency calls and route them to the right channels, dispatchers can give their full attention to situations that truly require their judgment, empathy, and experience. This shift directly impacts the quality of service and, consequently, the outcomes of emergency operations.
Another benefit observed in early deployments is a reduction in stress caused by the sheer volume of simultaneous calls. With AI helping prioritize and triage, dispatchers can work with more focus and less cognitive overload. In a field where burnout is an everyday reality and losing experienced professionals comes at an enormous cost, any measurable reduction in that level of pressure is extremely valuable.
What to Expect Going Forward
The trajectory of Artificial Intelligence in emergency dispatch services is still in its early stages, but the signs are very encouraging. As language models become more sophisticated and integration systems advance, the available tools are expected to become progressively more accurate, faster, and more adaptable to the local realities of each region.
Automatic translation, which already represents a significant leap forward, is expected to evolve to offer not just conversion between languages but also detection of dialects, regional slang, and cultural expressions that may be relevant to correctly interpreting an emergency. The more sensitive the system is to those kinds of nuances, the more useful it becomes for dispatchers who depend on that information to make quick decisions.
As Stephen Kennedy put it bluntly: 10 years from now, everybody is going to be using AI to make things more efficient. And judging by what is already happening in American dispatch centers, he is probably right.
The use of AI by emergency services reflects a broader trend we are seeing across society, from medical imaging analysis to fraud detection to autonomous vehicle navigation. The difference here is that when we are talking about 911, we are talking about a context where technology has the potential to literally save lives. And that completely changes the bar for what it means to do AI responsibly. 🚀
