It goes without saying that artificial intelligence (AI) is on everyone’s mind. We’ve all heard the narrative: AI is not optional; it’s not a buzzword; it’s going to affect every single industry. The outlook on AI is overwhelmingly positive from a business perspective, and this is the same case for monitoring centers.

Traditionally, the industry is slow to adopt new technologies, and often for good reasons, says Chris Brown, CEO, Immix, Charlotte, N.C. “Some of it is just fear and change,” he says. “The other thing is, we’re in the life safety business. It’s difficult to take big risks with how you operate your business every day. However, I am starting to see some companies take some big leaps. I think we’re closer to a rapid adoption of AI than we think we are.”

Why? The risk of escalation is higher today than it was 20 years. Because of this, operators are dealing with far more events, both critical and non-critical.

This is where AI comes in. Security industry leaders agree that AI is nowhere near ready to “take over” operations in the industry (and likely never will be). However, AI excels in handling repetitive tasks and non-critical events that the average operator wishes would just be automated so they can focus on the lives they’re committed to protecting.

In short, AI directly and effectively addresses “the people problem” that monitoring centers commonly struggle with: high turnover, unmanageable workloads, disjointed efficiency and grueling training processes.

AI Doesn’t Replace. It Augments.

AI is actively helping monitoring centers tackle these challenges. “AI and related automation technologies help address these issues by removing large amounts of routine work that can distract operators from the events that truly require their attention,” says Jim McMullen, president, COPS Monitoring, Williamstown, N.J.

These are widely considered the “low-hanging fruit” of what AI is currently capable of, meaning it’s not just available, it’s also accessible. “The widespread exposure to AI and all the dialogue and fast talk that it’s generating is creating more of a cultural shift,” says Chris Newhook, vice president of monitoring operations, American Alarm and Communications, Arlington, Mass. “People [customers and leadership] are driving the change and are more receptive and willing to embrace forms of automation that, while by no measure would classify as AI, nevertheless allow us to communicate by digital means.”

One thing AI cannot do, however, is replace people. “For us, this part is non-negotiable: AI supports the decision, but humans own the decision,” says Priya Serai, chief information officer, Zeus Fire & Security, Paoli, Pa. “Monitoring is a critical business. These are real people, real properties and real risks, so we design every AI workflow with one simple rule: AI can inform and accelerate, but it cannot decide.”

AI Is Your Next New Hire

Today, monitoring centers are using AI as an assistant to their skilled operators. Traditionally, operators had to memorize site layouts, response steps, customer preferences and device quirks. Now, tools like ZeusGPT, Zeus Fire & Security’s generative AI layer, surface the right information automatically. “The skillset shifts from ‘remember everything’ to ‘use the information well.’ This actually levels the playing field for newer operators,” Serai says. “When the job is less overwhelming, the talent pool opens up.”

Because of the amount of non-dispatchable, non-critical events — like a power outage — that come across an operators’ desk, it’s easy for burnout to occur. “It definitely takes a toll on people,” says Caroline Brown, CEO, Security Central, Statesville, N.C. The company built a conversational AI agent, Ella, to assist with these non-critical events. “Seeing that [Ella] is able to alleviate that need, and [operators] can focus on those critical emergencies, say a fire or a panic alarm, I think we’re starting to see a more positive interaction and a little bit lighter workforce because it is such a serious role when we’re triaging and helping these situations,” Brown says.

AI is also increasing operator efficiency by eliminating tasks that do not require a lot of skills or knowledge to complete. “We use AI-assisted video analysis to filter out irrelevant motion and highlight events that may require action,” McMullen says. “We use conversational IVR technology to handle routine outbound verification calls that do not require an operator. We use SMS and chat tools that enable subscribers to escalate real alarms or resolve non-emergency alarms quickly. Our goal is to remove distractions and routine tasks so our people can stay focused on emergencies and moments that truly require human intelligence.”

Guardian Protection, Warrendale, Pa., is implementing AI agent assist technology to support employees across a diverse range of tasks. “This strategic use of technology reinforces the value of human connection in service delivery, ensuring that while operational efficiency is optimized, the personal touch remains central to our customer experience,” says Jason Bradley, chief operating officer, Guardian Protection.

There are multiple ways AI addresses turnover, a particularly prominent industry issue. These solutions range from eliminating the parts of the job that are “boring” or “mundane” to catching early signs that someone is not fit for an operator role.

“A big thing that AI should help do is keep monitoring center staff engaged and excited,” Brown of Immix says. “Allowing operators to spend their time helping and winning for their customer keeps them excited, engaged and keeps them mentally healthier. I think that we’re going to find that people are more drawn to work in a center because it’s not a mundane, dark room to spend your day in. You’re actually doing things that are exciting and interesting, and you’re helping people. It will change the dynamic nature of who we get to hire and how people stay in the role.”

On the training front, he adds, “You’ll be able to tell AI, ‘Create me a scenario where there’s 20 people in a convenience store and someone comes in and robs it with a weapon,’ and run that scenario in front of an operator where they have to respond to it and take action to make decisions. I think we’re going to see AI-simulated threats become part of the training routine, and I think that will be incredibly helpful for the industry.”

Security Central has begun this process by building a learning management system (LMS) to make training more efficient. “When people aren’t necessarily talking on the phone to someone, especially in distress, we’re using AI to come up with stressful situations — they’re frustrated, they were woken up in the middle of the night, they don’t remember their passcode — and helping them through those things in a polite, professional [manner], but also with a sense of urgency,” Caroline Brown says. “We’re seeing … that we’re able to stress test [trainees] a little more. If someone is going to turn over rather quickly, we hope it’s in the training scenario.”

In the interim between training and becoming a veteran operator, Erin Bullard, director of partner relations, Immix, believes AI will also help with prioritization, ensuring that people who might not have the most training are not receiving the highest priority alarms or life safety threats that they are not prepared to handle yet.

Remaining Cautious & Maintaining Trust

We would be remiss to have a conversation about AI without discussing the ethical and legal considerations. Many monitoring centers are addressing these issues by forming departments dedicated to these technologies, carving out time in leadership meetings to discuss AI concerns, and over-emphasizing the need to remain human-centric.

Security Central’s quality assurance team, for example, is closely monitoring Ella’s accuracy and efficiency. The company’s leadership also has frequent conversations about the ethical uses of AI and actions they can take to ensure no information is being released to the wrong parties.

Similarly, American Alarm has an AI initiative team, and a number of the company’s senior and executive management personnel are involved. This is one of the first steps American Alarm has taken to apply controls with respect to sharing sensitive data related to company financial information. “It’s easy to look at all of these things that are coming our way and say, ‘Let’s just get on it, let’s get out there, let’s make sure we’re competitive, we’re leading edge.’ You’ve got to be very careful with personal information, company financial information, strategic plans,” Newhook cautions. “You can’t just cut and paste and drop that into a ChatGPT and say, ‘Give me the best approach for this.’ Right from the get-go, you want to circle the wagons to get all your people on the same page with respect to it.”

Serai and the Zeus Fire & Security team are firm in the belief that AI cannot be the final decision-maker, ever. “We are very intentional about this,” she says. “The ethical risk is obvious: if AI makes the wrong call and we dispatched (or didn’t dispatch), who owns that decision? To avoid that grey zone, we have a simple rule: AI can assist, but humans make the call. Always. This protects both safety and liability.”

COPS Monitoring has the same approach to and emphasis on the supporting role of AI paired with human oversight and decision-making. “Subscribers trust us with their lives and property, so we must preserve human verification and human decision-making at every critical step,” McMullen says. “We also take data privacy and security very seriously, especially in systems that record, analyze or store communication. Any AI tool that handles subscriber or event data must meet the strict security requirements of our SOC 2 certification.”

Bradley agrees. “Ultimately, our objective is to leverage AI as a strategic asset — augmenting our workforce, enhancing customer outcomes and upholding the highest standards of reliability and trust. This measured approach ensures that every advancement in AI is thoughtfully integrated.”

Brown of Immix is concerned with the legal aspect of using AI. “The challenge we’re going to find, which will get played out in courts probably very quickly, [is] when AI makes the wrong decision, who’s responsible? Is anybody responsible, number one? But who’s getting sued? Is it the monitoring center who chose that AI and put it to work? Is it the manufacturer of that AI who wrote it? Those are questions that have yet to be answered because we haven’t seen a lot of lawsuits filed and won or lost,” he says. Thus, he advises other leaders to be cautious and consultative when implementing AI solutions, and to not forget to add AI clauses to contracts, when applicable.

What’s Next?

AI is improving and developing at a pace that is nearly impossible to keep up with. It is just as difficult to predict where AI is going and what we’ll be able to do with it in the future.

“I caution everyone to not miss the fact that wherever you think we’re going to go in the next two years, we’re already there. It’s available,” Brown of Immix says.

He believes AI is headed in a direction that will allow operators to become much more predictive. “We’re starting to see large language models, or what’s kindly called LLM, in the space today,” he says. These tools will help an operator quickly look at historical data related to a particular alert, such as a black truck being parked in a fire lane a number of times within the week.

In the future, Serai sees a monitoring center amplified by AI. Operators are seeing pre-qualified events instead of endless feeds. An operator who’s only been on the floor for three weeks is working like they’ve been with the center for three years thanks to tools like ZeusGPT, which provides the site history, relevant signals, unique customer instruction, and the SOP highlighted step-by-step for any event. Supervisors aren’t chasing or correcting problems; they’re coaching and leading their people. New hires aren’t overwhelmed by screens and codes because they’re focused on scenario thinking. When an event pops up, AI automatically clips the video, identifies what’s going on, highlights the zone involved, and surfaces the SOP. “This isn’t science fiction,” Serai says. “This is where we’re heading: step by step, upgrade by upgrade, one operator win at a time.”

McMullen sees AI growing in areas such as video activity analysis, non-emergency triage and background task automation. “Routine and low-priority events will increasingly be resolved without an operator, and many support functions will become faster and more efficient. Operators will also benefit from real-time assistance that helps organize information, surface important details and reduce the workload created by routine tasks,” he says. “What will not change is the human-centered structure of a monitoring center. Emergency events still require intuition, empathy, communication and moral responsibility. These are the strengths that define human intelligence, and AI cannot replace them.”