Agentic AI in physical security: Redefining integration, innovation and impact

YARMOUTH, Maine — Agentic artificial intelligence (AI) is transforming the physical security landscape with its proactive and adaptive capabilities. This evolution is reshaping traditional roles across the industry, experts said during a recent webcast.

The webcast, “Agentic AI in Physical Security: Redefining Integration, Innovation, and Impact,” was hosted by Security Systems News and moderated by Editor Cory Harris. Panelists were:

• Don Morron, founder, HighlandTech
• Steve Reinharz, CEO, Artificial Intelligence Technology Solutions (AITX) and Robotic Assistance Devices (RAD)
• Steve Lindsey, CTO, LiveView Technologies (LVT)

How agentic AI is different

Unlike traditional AI, which depends on human input and predefined rules, agentic AI learns, adapts and acts independently. This autonomy is redefining roles for integrators, manufacturers and end users.

“The one word that derives from agentic is agency,” said Morron. “Another way to think about agency is autonomy, and I would say the difference between prior versions of AI systems in the physical security industry and these new agentic systems is the level of autonomy that the AI is given. You can look at it like a camera system that used object detection that was fixed, that was predefined (and) that is a form of artificial intelligence or deep learning. But this new generative AI era, what we’re now able to do is give these agentic AIs the agency to decide more of what’s happening in a given scene.”

Why agentic AI matters now

But why now? Harris asked panelists what’s driving the emergence of AI in the security space.

“Processing power,” Reinharz said. “We now have adequate processing power to do agentic AI, which requires a lot of power. We really needed chips to catch up with what we wanted to do, and we have that ability now.”

Lindsey highlighted the agility and dynamic nature of agentic AI, noting its ability to process vast amounts of unstructured data.

“It can reason through that very unstructured data that it’s never been able to do before,” he said. “So, we talk about Vision Language Models – there’s nothing more unstructured than that, right? The fact that it can break that down and transform it into data is a breakthrough that we haven’t really seen until just as of late.”

You can watch the full webcast online here.