Interviews
Lumana: How AI Is transforming video surveillance in banking
Global Banking and Finance Review
Company
For decades, banks have relied on traditional CCTV systems that acted mainly as passive recorders — useful for investigations after an incident, but offering limited real-time visibility into what was happening inside branches or around ATMs.
For decades, banks have relied on traditional CCTV systems that acted mainly as passive recorders — useful for investigations after an incident, but offering limited real-time visibility into what was happening inside branches or around ATMs.
AI is now transforming these once-static video systems into proactive intelligence tools that enhance security, improve operations and deliver real-time insights. This shift is redefining how financial institutions detect risks, streamline branch workflows and protect customer trust.
Global Banking & Finance Review recently spoke with Ofir Mulla, Cofounder and CTO at AI video security provider Lumana, to learn more about what banks and financial institutions should know about how AI is transforming video surveillance.
How have banks and financial institutions traditionally used video surveillance in their branches and physical operations?
Banks have historically relied on CCTV as a deterrent and as a record for post-incident review. Cameras were placed over teller lines, vaults, ATMs, entrances and parking lots, but the systems were largely passive — footage was only reviewed after something went wrong. Security teams often had to manually sift through hours of video and integrations with alarms or access control were limited or inconsistent. As a result, surveillance was useful for investigations, but offered little in the way of real-time awareness or operational value.
What new capabilities does AI bring to traditional CCTV and security camera systems in the branch environment?
AI has transformed traditional cameras from simple recorders into real-time intelligence tools. Branch teams can detect behaviors, such as loitering near ATMs, unusual movement patterns or prolonged inactivity at workstations, which may signal risks. Beyond security, AI can provide operational insights by tracking queue lengths and other operational bottlenecks inside the branch. Managers can use this data to optimize layouts, staffing levels and other efficiencies across locations.
AI solutions like Lumana can also provide early indications of risk that human observers often miss. For example, Lumana can detect ATM tampering, identify confrontational behavior that could escalate, and flag unauthorized after-hours access or policy violations in sensitive areas. It can also detect safety issues such as smoke, slip-and-fall incidents, or sudden crowding. And operationally, it highlights long wait times, understaffed periods, or inefficient branch layouts, allowing banks to act before problems affect customers or escalate into losses.
When AI models are trained on sensitive footage involving customers or employees, what privacy, compliance or data-governance obligations must banks consider?
Video footage in banking environments is regulated personal data, so governance must be built in from the start. Banks should restrict usage to defined purposes, enforce strict access controls and ensure clear audit trails for every workflow. Technically, banks should limit exposure by processing video at the edge, when possible, and maintaining clear documentation of model training sources, updates, and performance checks. Compliance teams must ensure lawful bases for data capture and provide transparent privacy notices to customers and employees to meet regulatory obligations and uphold trust.
How should banks evaluate whether their existing camera infrastructure is ready for AI analytics or if upgrades are necessary?
Most modern camera networks can support AI, especially with edge computing. Banks should assess image quality, camera angles, and frame rates to ensure the footage is suitable for accurate analytics. It’s also important to evaluate network capacity and bandwidth constraints.
The best approach is to start with a small pilot at a branch or ATM cluster to validate performance, identify gaps, and make adjustments before scaling. This allows institutions to gain quick wins without major upfront investment.
Where does Lumana fit into this ecosystem for AI-driven video surveillance?Lumana is a hybrid-cloud AI video surveillance system designed to work with the cameras banks already have. Instead of requiring full replacements, we can transform existing cameras into intelligent devices, offering centralized management, real-time alerts for specific activities and behaviors, fast investigations, and AI dashboard capabilities that turn video data into actionable insights.
AI is processed at the edge, providing the low-latency awareness banks need, while strengthening privacy and minimizing data movement. For financial institutions balancing security, compliance, and operational efficiency, Lumana provides a scalable pathway to advanced AI capabilities without disrupting existing systems, elevating visibility, accelerating response times, and meeting regulatory expectations, all while leveraging the infrastructure already in place.
What measurable business or operational benefits beyond security can banks achieve by modernizing their video surveillance with AI?
AI-powered video security systems dramatically reduce investigation time from hours to minutes by allowing staff to search footage using natural language or specific attributes. Banks can detect ATM or teller manipulation earlier, helping reduce fraud losses and operational shrinkage. AI-derived insights into queue times and customer flow improve service quality and workforce allocation are additional examples. Enhanced incident capture also reduces liability exposure and supports more efficient insurance claims. Institutions that adopt AI consistently report faster response times and measurable improvements in fraud detection and branch efficiency.
What emerging trends in AI video surveillance should banks prepare for over the next two to three years?
AI models will increasingly adapt to the unique environment of each branch, reducing false positives and improving accuracy over time. Video analytics will merge with transaction data, access control logs and fraud-monitoring systems to uncover more complex patterns of criminal behavior. Privacy regulations will continue to evolve, making anonymization, on-device processing and federated learning standard expectations. Operational decisions, such as staffing, layout adjustments or customer-flow design, will increasingly be informed by continuous video-derived insights.
Any final thoughts or recommendations for banks considering an upgrade to their video surveillance systems?
Start with clear goals and not a list of technologies. Identify one or two high-value use cases, such as tampering detection or queue monitoring, and launch a small pilot using existing cameras plus edge hardware. Build privacy and governance into the system from the outset so it scales responsibly. And ensure the video platform integrates into fraud, operations, and security workflows rather than operating in isolation.
The institutions that measure outcomes — faster investigations, fewer incidents, shorter queues — will build the strongest business case and scale most effectively.
Ofir Mulla, Cofounder and CTO , Lumana
Explore more articles in the Interviews category
