Many retailers face challenges such as shrinkage, organized retail crime (ORC), and staffing shortages, so they’re looking for solutions that enable them to do more with less. Leveraging more intelligent security systems and artificial intelligence (AI) technology can be a force multiplier. These solutions can improve efficiency, enabling retailers to address security challenges more effectively while enhancing operations and the overall customer experience.
In the 2025 Genetec State of Physical Security Report, 78% of retailers noted that they’re using AI to trigger events of interest. The following three top objectives included improving emergency response dispatch, automating repetitive tasks, and filtering events for assessment.
Most likely, some form of AI processing is embedded in the software you use daily. Such technology can detect patterns, classify data, and deliver analytic insights. Retail security teams can automate and streamline workflows, enabling them to respond quickly to business disruptions. Used responsibly, AI-powered analytics can significantly enhance security and improve operations.
AI Versus IA
The emergence of generative AI, the technology used in tools such as ChatGPT, has triggered a new wave of AI buzz within multiple industries, including retail. While there is a lot of hype, it also has the potential to significantly change how security systems are deployed, configured, and utilized.
Before we discuss use cases, let’s clarify some of the terminology. AI is a broad term encompassing a range of technologies. It refers to technology and processes that enable machines to learn from experience and adjust to new situations without explicit programming.
AI-enabled applications can help retailers achieve specific business outcomes that heighten productivity, security, and safety. However, it’s essential to realize that AI is a technology, a set of tools.
The outcomes that the technology delivers are more important than the tool. These outcomes are part of a comprehensive approach known as Intelligent Automation (IA). IA encompasses AI combined with intuitive user experience (UX) design, rich data models, and rules and policies, all while maintaining a human-centric approach.
Let’s break this down:
- Automation is when tasks, whether easy or hard, are done without people needing to do them. Once a process is set up in a program, it can be repeated as needed.
- Traditional automation requires a clear definition from the start. Every aspect, from input to output, must be carefully planned and outlined by a person.
- Intelligent automation (IA) enables machines to handle both complex and straightforward processes without requiring explicit instructions. IA typically utilizes machine learning, such as generative AI and natural language processing, to suggest ways to analyze data or take actions based on existing data and usage patterns.
These solutions empower security operators with the correct information at the right time, ensuring they can focus on core activities instead of searching for and analyzing data patterns.
AI-Enabled Technology to Enhance Retail Security
To address ORC, retail crime, and shrink, today’s retail loss prevention investigators may be tasked with reviewing hundreds of incidents at different stores. They’re also trying to prevent internal theft at warehouses and distribution centers. Strengthening processes and introducing better monitoring systems can help retailers become more resilient to these threats.
Let’s explore a few ways retailers use IA to improve their safety and security strategies.
1. Empowering loss prevention teams by reducing friction -- When cameras record during store hours, retailers collect hundreds of hours of video per store every week. That’s a lot of data to sift through. Simply gathering and storing that video footage isn’t enough. Retailers are equipping their loss prevention teams with tools to use the footage efficiently, potentially proactively, to address threats.
Rules and actions established within the security system can help reduce false alarms and alleviate the strain of monitoring hundreds of cameras. Automated alerts notify operators of relevant video indicating a potential threat or incident. By utilizing these computerized functions, security teams can stay focused on what’s important and avoid the need to monitor all cameras simultaneously.
For example, a security system can alert operators to potential losses. This may include an individual or group spending long periods unsupervised in a high-value aisle. Once the alert is sent to the security team, management can determine whether the area needs extra monitoring or attention.
2. Speeding up investigations -- AI-enabled technology can also assist in post-incident investigations. Manual searches of video footage can take hours and even days, sapping resources and slowing investigations. Advanced system automation and IA can help bridge the gap between response and investigation.
Instead of scouring hours of video to find a specific clip, IA allows investigators to do so within minutes. Modern security systems feature natural language forensic search capabilities, enabling operators to locate the necessary footage quickly. By simply providing a prompt, such as “man in a blue shirt” or “red truck with a box,” the operator can easily view all video clips matching that search within a specific time.
Once the footage is located, loss prevention teams can use a digital evidence management system (DEMS) to quickly and securely share the evidence with law enforcement. Every step is done digitally, eliminating the hassle of physical copies and making collaboration between retailers and law enforcement faster and easier.
3. Detecting vehicle license plates -- Organized retail crime groups often target stores in specific areas. A network of retailers in a community can install automatic license plate recognition (ALPR) cameras across their locations to collectively capture data. This means each location can track vehicle license plates that enter their premises, and access shared ALPR data with other participating retailers nearby.
When a vehicle tag linked to previous thefts is detected at a location, a real-time notification is automatically sent to the store where it is flagged. By informing partner stores about the suspect, the retailers can collaboratively prevent further thefts, build stronger cases against offenders, and ensure better asset protection.
More Than Security: Improving the Customer Experience
While modern security platforms with AI-enabled technology can significantly enhance security and investigations, these solutions also offer valuable insights for improved operational efficiency. Retailers can correlate video with other systems to gain insights into the customer experience. Cross-referencing video footage with heat mapping, people counting, and other analytics can help retailers understand the customer’s journey through the store and at checkout.
This insight provides opportunities to enhance the customer experience, boost sales, and more effectively manage staffing. Here are four ways retailers are deploying IA to improve the customer experience.
1. Understanding customer flow -- Using machine learning, retailers can analyze data from in-store cameras to gain insight into foot traffic patterns and their variations across different days and times. This makes it easier to establish data-driven benchmarks for customer wait times for protected products and at the point of sale (POS). They can more effectively redistribute staff where needed most to provide a better shopping experience and greater sales.
A large coffee store chain implemented advanced analytics to study the customer flow and determine the optimal store layout. Where is the best location for self-serve items, such as straws, napkins, and sugar? What signage is needed to direct traffic? Answers to these questions helped them improve their story layouts and the overall customer experience.
2. Boosting sales by rethinking displays, marketing, and promotions -- People counting and direction of movement analytics can help retailers understand the effectiveness of store displays. Security system analytics can track the number of people who view the display. Retailers can then compare that to POS data on the number of products sold. Calculating the conversion rate of displays gives concrete feedback to evaluate which approach works best.
Sales and marketing departments can also gain insights into the success of their initiatives by tracking store traffic patterns before, during, and after a promotional campaign. They can draw annual comparisons and identify which programs drove traffic, down to the day of the week and time of day.
3. Prioritizing restocking and cleaning -- Retailers are getting creative about using video data to optimize operations. Video data can be used to keep track of empty shelves, making restocking and reordering processes more efficient.
Video analytics can also help address store cleanliness and notify staff if there are signs of disorder, such as a toppled display or stock blocking a store aisle or exit. Some stores are deploying cameras to monitor shelves and specific areas of their retail space.
4. Queue detection -- A long line detected at a register can signal an alert to management. Supervisors can then reduce bottlenecks by opening up a new cash register. The response creates a positive customer experience and may lead to increased transactions.
Management can use the technology beyond the checkout areas. By monitoring zones throughout the store, they can alert team members to customers who may need assistance, allowing an employee to help close a prospective sale.
Embracing Intelligent Automation with a Human in the Loop
For most retailers, implementing AI involves a couple of driving factors: achieving large-scale data analyses and higher levels of automation. They aim to leverage their physical security investments and data to enhance the effectiveness of their loss prevention teams.
Even with these solutions, it’s essential to understand that most AI solutions in physical security aren’t one-size-fits-all. While AI is a key component to reaching higher levels of automation, much consideration, forethought, and planning are still required to achieve accurate results.
It’s essential to be cautious when exploring new AI solutions. AI models can’t make critical decisions on their own. When selecting new software, consider a few key best practices to keep in mind as technologies continue to evolve.
- Keep data protection and privacy in mind: The manufacturer should only use datasets that respect relevant data protection regulations. Likewise, your system must include robust authorization and authentication measures to ensure that only authorized individuals have access to sensitive data across your AI-driven applications.
- Focus on transparency and fairness: Look for manufacturers that rigorously test their AI models to ensure they deliver balanced, equitable, and bias-free results. The results should also be accurate and explainable. Make sure humans have the final say: Machines can’t grasp the intricacies of real-life events like a security operator can. AI models present data in a way that helps humans make more informed choices, but humans should always be the decision-makers.
How to Capitalize on AI in Physical Security
One of the most effective ways to leverage new AI advancements in physical security is to implement an open-architecture security platform. Open architecture allows security professionals to explore applications that drive greater value across their operations.
AI is a tool that has the potential to alleviate operators' burdens by providing the correct information at the right time. Retailers who adopt an open-architecture system will be in the best position to adopt new technologies as they are introduced. With an open-architecture unified platform, you can leverage solutions that work for you now while monitoring developments that deliver better, more cost-effective security in the future.
