Evaluating AI in Security: 13 Questions to Ask at ISC West

Each year ISC West reveals how rapidly security technology is evolving. AI continues to transform nearly every aspect of the security industry, driven by companies investing heavily in AI-enabled products. Security and building management systems predominantly employ narrow AI, known for its reliability and low error rates compared with general or generative AI.

However, for physical security applications, narrowly focused generative AI can play a critical role in translating real-time data into actionable, scenario-based situational awareness explanations.

Additionally, increasing connectivity and computing power elevate the cybersecurity risks, making Zero Trust architectures essential.

Platform-level AI, intelligent system and device integrations and system-of-systems architectures are transforming what for more than 50 years has been a device-centric industry into highly scalable, real-time situational awareness infrastructure. These architectures increasingly depend on interoperability between systems from multiple vendors within the overall security operational architecture. As a result, open standards and vendor-neutral integration approaches are becoming important design considerations for future-ready security deployments.

For a clear picture of how these changes affect security system design and operations — and how they may reshape your thinking about what your security capabilities are today versus what you’ve wanted them to be — see my articles:

• Four-Dimensional Security Surveillance
• Physical Security Watershed Moment

Leading companies such as Ambient AI (23005), Allied Universal Services (27099), Brivo Systems (20031), Cobalt AI (9123), Eagle Eye Networks(20037), IQSIGHT [formerly Bosch Video Systems] (11053) and Milestone Systems (18053) — which has adopted the ethical G7 Code of Conduct for AI — and many others continue to significantly expand their AI capabilities.

High ROI perspective

Walking the ISC West exhibit floor can be exhilarating — and overwhelming. New AI-powered cameras, analytics engines, drones, robotics platforms and sensors promise dramatic improvements in security effectiveness. But evaluating these technologies individually can obscure an important reality: the greatest operational value often comes from how well they work together.

Artificial intelligence will again dominate many conversations and product demonstrations at the show. In recent years, however, the discussion has begun to shift. Rather than focusing only on individual AI-enabled devices or analytics features, security professionals are increasingly evaluating how technologies interact to improve operational outcomes.

Many emerging technologies — AI analytics, autonomous drones, robotics, advanced sensors and situational awareness platforms — deliver their greatest value when integrated into a larger operational architecture. Increasingly, enterprise physical security deployments resemble what the National Institute of Standards and Technology (NIST) describes as cyber-physical systems — interacting networks of devices, software platforms and human operators functioning together as a coordinated system of systems.

Evaluating technologies within that broader context can significantly improve both operational effectiveness and return on investment. A drone platform, for example, becomes far more valuable when integrated with perimeter detection systems, AI video analytics and security operations workflows.

Another example of innovative sensing technology comes from Acoem (Stand: 28039). Drawing on more than three decades of acoustic engineering expertise, Acoem has developed an AI-enabled acoustic sensor capable of detecting gunshots, sniper events, explosions, firecrackers and vehicle accidents. Each sensor can monitor areas up to roughly a 500-foot radius and can trigger automated PTZ camera slewing and audio replay within seconds of detection.

Like many advanced sensing technologies, however, its greatest value emerges when integrated into a larger automated response workflow that informs security personnel and other stakeholders with the appropriate situational context — such as that provided by platforms like Cobalt AI.

Architectural lens for evaluating technologies

Many technologies on the show floor are presented as standalone capabilities. In practice, their greatest value is typically achieved when they operate as part of a larger system-of-systems architecture.

Evaluating technologies and designing systems through this architectural lens — how well they integrate with other systems, contribute to situational awareness and support coordinated detection, verification and response — can significantly improve operational effectiveness and long-term return on investment.