Artificial intelligence (AI) is reshaping the security industry, driving significant advancements in technologies like video surveillance, access control and video analytics. For systems integrators, the rapid pace of AI adoption presents both opportunities and challenges, from navigating complex implementations to addressing the high cost of these cutting-edge solutions.

Ahead, we examine AI’s impact on the security landscape, focusing on how integrators assess their clients’ needs and effectively communicate AI’s value while managing budget constraints. Insights from interviews with leading integrators reveal their experiences with AI integration and its implications for their businesses and end-user clients.

Additionally, this report highlights how integrators are utilizing AI internally to enhance operational efficiencies and customer service. Perspectives from manufacturers further illustrate how they support integrators in preparing for AI integration through comprehensive training programs and resources.

AI’s Significant Impact on Surveillance

While multiple security technologies are being disrupted by AI and its related subsets, video surveillance stands out as the most intensely impacted, according to sources interviewed for this story. Integrators are finding that AI’s advancements in real-time analytics and threat detection are driving significant changes in how these systems are designed and deployed, presenting both opportunitiesand challenges.

“AI is truly transforming the security landscape, and we’re seeing the biggest impact in video surveillance,” says Priya Serai, chief information officer, Zeus Fire and Security, Paoli, Pa. “Traditional systems that relied on manual monitoring are evolving into intelligent, proactive tools. With AI, video surveillance has shifted from just passively recording footage to real-time analysis — identifying suspicious behavior, detecting unusual patterns, and even predicting potential security breaches before they occur.”

Access control systems are also seeing significant transformation, explains Serai, who is featured on this month’s cover. AI-driven biometrics, facial recognition and smart access patterns make these systems not just more secure but more seamless for users.

“And we can’t forget intrusion detection — AI’s ability to differentiate between real threats and false alarms is helping teams focus on what matters and respond faster,” Serai continues. “At the end of the day, AI is all about converting data into actionable insights, and that’s what’s making these systems smarter, faster, and more effective than ever before.”

At Chimera Integrations in Syracuse, N.Y., the 2024 SDM Systems Integrator of the Year, video surveillance has experienced the most significant transformation due to AI, says Vice President Justin Stearns. He explains that AI’s ability to provide real-time threat detection and pattern recognition has greatly enhanced the effectiveness and adaptability of these systems.

“Intelligent surveillance is replacing card readers for access control, traditional IDs,” Stearns adds. “This is a guess, but I think surveillance equipment could potentially replace fire alarm systems in the not too distant future.”

Amir Shechter, executive director of innovation and technology at Convergint, Schaumberg, Ill., notes that while AI transformation in the security industry is still in its early stages, the landscape is evolving rapidly. He explains that AI’s performance, accuracy, scalability and use cases are steadily improving, not just in video applications — where object detection and classification generate metadata or alerts — but also across a range of emerging applications, including:

Access control systems, which prioritize dynamic data management, insights, authentication and workstream logic.

Smart sensors, such as light detection and ranging (LiDAR) and radar to classify, track and detect objects.

Orchestration software such as PSIM and PIAM to incorporate data insights, automatic report creation, language based queries, dynamic event standard operating procedures (SOPs), anomaly detection, smart recommendations, etc.

Moreover, Shechter explains, challenges such as labor shortages, regulatory demands (e.g., alarm verification), and the high costs of employing security guards are driving the demand for increased automation and cost control.

“These pressures present a significant opportunity for the market to introduce new capabilities and services that deliver real value to customers, such as AI-driven remote alarm monitoring,” he says. “Finally, the shift toward cloud-friendly architectures is enabling innovations that were previously constrained, facilitating and expanding the potential for new offerings.”

Prosegur, Deerfield Beach, Fla., is another systems integrator witnessing a surge in video surveillance offerings, as end users increasingly expect their existing systems to deliver more through advanced analytics. Mark Barbaric, senior vice president of technology integration at the company, notes that analytics once focused mainly on security features like cross-line detection, movement tracking, object left behind and suspicious behavior. Now, new analytics offer valuable business insights, such as customer wait times, spill detection, personal protective equipment (PPE) compliance, and facial recognition for access control — often using something as simple as a standard iPad camera.

“By enhancing existing video surveillance infrastructure with cameras that incorporate AI capabilities, many specialized IoT sensors are becoming redundant, streamlining design and maximizing the most use of the camera,” Barbaric explains, but adds, “This will not negate the need for special sensors in the near term; some use cases are very specific and require purpose-built devices.”

How AI Has Changed Systems Design

As AI continues to reshape the security landscape, security integrators are finding that the design and installation of various systems have been significantly influenced by these advancements. From modifying system architecture, to incorporating AI-driven analytics, to rethinking hardware configurations for better compatibility, the impact of AI is prompting new approaches. Integrators are adapting their methods to meet the demand for more intelligent, efficient and scalable solutions that leverage AI’s capabilities to deliver greater value for end users.

“AI has transformed system design and installation in the physical security industry, by going beyond traditional access control and video management solutions,” Shechter explains. “Its focus is on solving inefficiencies and offering new data insights. The design process begins by understanding the problem, defining objectives, and evaluating the data needed for informed decision-making.

Choosing the appropriate AI model is essential, followed by rigorous testing, validation and fine-tuning, Shechter says. To achieve peak performance, AI systems, like sensor placements, must be carefully integrated with hardware and software. Continuous monitoring and adjustments are necessary to ensure these systems evolve over time, delivering lasting value.

“AI solutions are not ‘set it and forget it’ applications,” Shechter adds. “They require ongoing attention, including a comprehensive approach that connects hardware, software, data, logic and integration. Once installed, the system must be continuously monitored and improved, ensuring that the AI model evolves with changing data and needs. This continuous cycle of refinement ensures that AI delivers sustained value over time.”

AI-based solutions have fundamentally transformed Chimera Integrations’ approach to system design, whether or not the planned installation includes advanced analytics or AI, explains Stearns.

“Historically we considered surveillance coverage from a forensic perspective,” he says. “What coverage does the client need to have supervision over their organizations areas of concern? Now the question is what problems are you having in your facility?”

For example, is PPE required and a concern for the end customer’s facility if not worn properly? Will automating the workflows associated with the supervision of PPE benefit the organization?

“We can now create proactive solutions that can stop something from happening,” Stearns says. “That is a complete change in our engineering principals.”

Stearns cites a specific use case example: A customer sought a glass-break system, self-monitored by security staff, for their offices with external windows. Instead, for half the cost, Chimera installed two 360-degree cameras with machine learning tied to PoE speakers outside the building. The integrator programmed line-crossing analytics, creating a 20-foot perimeter barrier. If crossed during specific hours, an automated message played, alerting security staff, who could then communicate with the intruder using two-way audio through the speakers.

“Cameras wouldn’t have been useful here prior to advancements in machine learning,” Stearns says. “The solution we went with will also alert on anomalies it detects in its environment.”

Barbaric explains that traditionally, Prosegur’s installation approach focused on designing security systems to protect people, property and reputation. The conversation has now evolved to consider how video cameras can serve multiple operational purposes beyond safety and security. Nowadays, discussions with end customers start by explaining the readily available analytics with current technology, then proceed to develop a proof-of-concept design that meets most customer expectations within a reasonable budget.

“Other analytics that may require additional or replacement hardware are then considered as optional, as the costs of supply and installation start to become more significant,” Barbaric says. “This strategic approach allows us to maximize the value of existing video surveillance in various business operations without a major capital cost upgrade to the end user.”

Serai emphasizes that AI is fundamentally transforming how Zeus designs and installs security systems, taking the company far beyond merely setting up cameras or access controls. Today, the integrator focuses on designing systems that are not only secure but also intelligent, scalable, and adaptable to evolving needs.

“Take video surveillance, for example. With AI-driven video analytics, camera placement is more strategic than ever,” she explains. “Instead of simply covering an area of concern for a customer, we also focus on key zones to leverage AI’s capabilities to detect behaviors, recognize faces or identify objects in real time. It’s all about placing cameras where they’ll generate the most valuable data for the customer and working closely with customer to ensure those insights are actionable and tailored to their environment. It’s smarter placement, not just more coverage.”

For access control systems, AI allows Zeus to integrate biometric authentication and pattern recognition seamlessly. This enhances security while making access more frictionless for users. Serai says the company is designing systems that can set access rules based on behavior patterns, providing a more responsive and user-friendly experience.

“We’re also exploring how to make installations more efficient and expect to have systems that automatically calibrate, learn their environment and self-optimize, leading to less manual intervention, more predictive maintenance and smarter updates based on actual usage,” she says.

Similar to other forward-thinking integrators, AI’s impact extends beyond security at Zeus. Serai describes how the company is leveraging operational analytics to unlock video data’s potential, offering insights into both security and business operations. From understanding customer flow in retail to optimizing warehouse processes, video intelligence is helping clients make better decisions.

“AI is pushing us to rethink security design,” she says. “It’s not just about hardware and coverage anymore; it’s about data, intelligence and delivering greater value that evolves with our customers’ needs.”

Effective AI Solution Implementation

When it comes to successfully implementing AI solutions, security integrators must first thoroughly evaluate their clients’ specific needs to determine which AI technologies can offer the most value. From assessing operational challenges to aligning AI with industry-specific use cases, integrators play a crucial role in selecting the right solutions and demonstrating the tangible benefits AI can provide, ensuring that each deployment enhances security and operational efficiency for the end user.

“We start by taking AI out of the picture and instead focus on the problem you want to solve,” says Satish Raj, CTO for San Antonio-based Pro-Vigil, a provider of AI-enabled remote video monitoring solutions. “We’re focused on having a workflow in which we can have accurate crime prevention. We started with that baseline and, once you set your accuracy needs, ask yourself how do you satisfy customer needs at this service level? And then how do you scale this up to thousands of customers? For us, AI is a force multiplier that allows us to offer that quality of service at scale.”

When deciding if AI is a fit for individual customer sites, Pro-Vigil looks at certain criteria including industry type. An outdoor construction site, for example, has different needs than a car dealership. “The same goes for a site that is fenced or gated — what we would recognize as a threat would be different there versus a site that is open to traffic,” Raj says. “From there, we create a customer profile and then we can apply AI to meet specific needs.”

To determine if AI-enabled solutions fit a client’s security needs, it’s crucial to ensure clear benefits like ease of use, streamlined operations and enhanced support, Shechter of Convergint explains. The key challenge is delivering value that justifies the investment, with a strong focus on ROI to drive widespread adoption beyond the proof of concept.

“This is why upfront consulting, advisory services and technical innovation are essential. They help ensure organizational alignment, stakeholder buy-in and a willingness to manage change,” Shechter says. “Establishing realistic goals, achievable technical parameters, a phased approach, and the assignment of the right resources are all key. Ultimately, creating a collaborative project environment that includes internal and external stakeholders — such as the customer, integrator and solution providers — ensures the project’s success.”

Stearns of Chimera Integrations also emphasizes the necessity for collaboration. Initially, the technology is vetted in-house before being considered for a client. For larger installations, a demo is set up in the client’s facility to allow the technology to demonstrate its capabilities. For example, Chimera recently designed a 250-camera system for a manufacturing facility. The primary concerns for the solution were employee safety and property damage caused by forklift accidents. A demo camera from a large, well-established manufacturer with previously vetted AI solutions was installed.

“The manufacturer supported us with the demo installation and programming and ultimately the solution failed to detect or prevent the situations we thought it would,” Stearns describes. “Ceiling height in this case caused the analytics/AI not to work without a significant increase in camera resolution that would have made the project go out of budget. We did an about face and installed a second demo with a different camera manufacturer and closed the sale on the project.”

Assessing AI-enabled solutions for a client’s security needs involves understanding their unique environment, pain points and goals, says Serai of Zeus, which takes a consultative approach to ensure AI adds true value. The first step is evaluating the client’s specific security challenges and desired outcomes, such as real-time incident detection, access control enhancements or operational insights from video data.

The environment and context — including facility layout, foot traffic and existing infrastructure — are crucial for the effectiveness of AI solutions. For example, Serai says, optimizing customer flow in retail spaces involves considering both security and operational analytics to improve sales, staffing or inventory placement.

Scalability and integration are key factors as well, ensuring the AI solution can adapt to evolving needs and integrate with existing systems. Flexibility is essential for long-term value and ease of expansion. Additionally, the client’s readiness for AI, including infrastructure and willingness to adopt a data-driven approach, is considered. AI-enabled security requires continuous learning, refinement, and a shift in how teams use and respond to data, she advises.

“Ultimately, it’s about finding the right balance — the AI solution should enhance security, provide actionable insights, and align with the client’s overall business goals,” Serai says. “At Zeus, our focus is on ensuring the solution not only fits but truly makes an impact on protecting people, property and profits.”

Overcoming Cost & Budget Constraints

Security integrators often encounter financial challenges when implementing AI solutions, primarily due to the high upfront costs involved. However, integrators are finding ways to demonstrate the long-term value and return on investment (ROI) of these systems to their clients. By focusing on the operational efficiencies and enhanced security that AI can deliver over time, they work to balance immediate financial concerns with the broader benefits AI brings to end users.

Many security departments excel in operational management but struggle with the business aspect of security, Shechter explains. Building a strong business case and financial rationale for AI integration can be challenging. It involves clearly defining the specific needs, value and financial advantages that AI offers, as well as creating a detailed cost-benefit analysis.

“Clients may be hesitant to adopt AI due to skepticism from previous digital transformation initiatives that under-delivered,” Shechter says. “Without a solid, realistic plan that assures success, this skepticism can become a significant financial barrier, discouraging investment in AI solutions.”

In the physical security sector, the high initial costs of AI systems can be offset by extending their capabilities beyond security functions, Shechter says. For instance, AI-enabled cameras can also be used to enhance safety protocols, manage logistics and gather occupancy data that aids in more informed decision-making across various operations, adding value beyond security alone.

“AI tools can help manage people’s movement, monitor hazards, enhance employee and guest engagement, and contribute to ESG initiatives,” he adds. “By collaborating with other business units, organizations can maximize ROI and tap into additional budgets, making AI adoption more feasible and aligning with overall business goals.”

With ever-evolving technology, Barbaric says Prosegur is finding that basic AI functionality is becoming more and more standard in modern hardware and the software that supports it. For customers that are on a single video and access control platform, the upfront costs are negligible, he says, while long-term ROI can easily be demonstrated for upgrade to more advanced systems.

“Most of our customers recognize that challenge but have difficulty convincing the C-level to spend limited budgets on major upgrades where technology could rapidly become outdated,” he says. “Our approach is to work with manufacturers to design a hardware solution that is future-proofed and open to various software applications, should the need arise in the future for advanced AI projects that deliver high ROI.”

For Pro-Vigil, building AI models in-house has been a tremendous benefit for the company and its customers; however, it took years of expertise and investment to establish this infrastructure, Raj explains.

“Because of our scale of operation, we process tens of millions of events every month. Therefore we are in a unique position to use outcomes from this processing to train AI on an ongoing basis,” he says. “These are all benefits that are ultimately passed on to the customer in the form of faster, more accurate crime detection and deterrence.”

Balancing the high upfront costs of AI systems with their long-term value and ROI requires a clear strategy, Raj says. This includes implementing approaches that reduce the time it takes for AI-driven benefits to materialize, ensuring that the investment provides measurable returns sooner rather than later. If the benefits are vague and take too long to materialize, it might not be the appropriate solution.

“Having a defined roadmap helps solve the problem faster,” Raj continues. “If you start with a well-defined pain point, you can translate that into the value proposition and ask, ‘Am I willing to pay the upfront cost to solve the pain point?’ For Pro-Vigil, that pain point was the ability to scale. As our company grew, we were not able to scale with the same accuracy and that’s when we brought in AI several years ago.”

According to Serai, Zeus is focused on balancing AI’s high upfront costs with long-term ROI by maintaining transparency from the outset, ensuring clients understand both the initial investment and the potential value over time. The company highlights the likely savings and efficiencies, enabling clients to see the bigger picture.

To ease the financial burden, Zeus often starts with smaller use cases that deliver “quick wins,” showcasing immediate ROI. Scalability is achieved through phased approaches, beginning with pilot projects and expanding as results become evident. Zeus also emphasizes AI’s adaptability, showing clients how systems can improve and evolve, delivering increasing value as AI learns.

“Ultimately, it’s about showing a path from upfront investment to long-term gain while ensuring that each step is aligned with the client’s needs,” Serai says. “We don’t have all the answers yet, but we’re dedicated to finding that balance and making sure our AI solutions deliver growing value over time.”

Stearns notes that while the benefits of AI generally outweigh any additional costs, the biggest resistance typically comes from subscription fees, which clients often question. However, this pushback isn’t directly related to AI being part of the solution itself, he says, but is more of a broader concern regarding ongoing costs. He emphasizes that if the AI solution wasn’t financially viable for the client, Chimera wouldn’t have offered it in the first place.

“We also typically have a good, better and best solution for any technology we are presenting, so if budget is a concern the client can decide what features they aren’t willing to pay for,” Stearns adds.

Stearns challenges the assumption that AI systems always come with higher upfront costs. He explains that Chimera often designs systems using both AI and non-AI solutions, and in their experience, AI-based systems don’t necessarily have greater initial costs compared to alternatives. He emphasizes that this cost differential depends on the specific system and application rather than being a universal truth.

“One thing that saves money on installation is we don’t have to do as much programming for the cameras and typically the UI is easier to use, so we save on end-user training costs,” he explains. “In many cases, we don’t need to install as many cameras when using an AI system versus a traditional VMS that lacks these features. Furthermore, a lot of the VMS companies have HaaS (hardware as a service) models for their products, so now there is very little CapEx expenditure and the system is more budgetable and stays in their OpEx budget.”