AI connects technology with security forces to enhance readiness
The CEO of an international marketing firm once told her audience at an industry symposium several years ago that we are all the custodians of data and we must let clients know it’s safe in our hands. She stated: “But with knowledge comes power, and with power comes great responsibility.” And as the evolution of advanced artificial intelligence (AI) platforms begin to permeate the security and risk landscapes, understanding how to integrate potentially unwieldy metadata into knowledge that eventually intersects with human operators is no small task.
There is no starker example of this paradigm shift than security environments like manned guard forces that, until recently, have been seldom touched by advanced technology. Yet today, we are seeing radical change in this once staid sector. New technology is now helping organizations better integrate advanced workflow automation, robust data capture and visualization and artificial intelligence into an ecosystem that drives better risk outcomes.
Making AI “Smart” With Humans
When your foundation has been built on security and facility services and spearheaded by manned guard personnel and a burgeoning array of data-driven software solutions that support their mission and those of myriad vertical market clients, it is difficult to separate man from technology. Last week at Allied Universal, one of North America’s largest and most progressive security services and technology companies, the release of an advanced artificial intelligence platform called HELIAUS, designed to improve safety and reduce risk by enhancing on-site guarding services, signaled a major innovation for security officers and their ability to detected and respond.
“One of the big challenges that we're taking on with HELIAUS, and in particular with the AI that we've employed, is to make it work efficiently with people,” explains Mark Mullison, CIO for Allied Universal. “This is a big challenge. We don't want our system to be smart like humans. We want it to be smart with humans.”
For Mullison and his team, the ability to orchestrate interactions between people and security technology like cameras, and drones, and access control systems, and various other IoT sensors is absolutely critical. He realizes that in the future, the walls between physical security and cybersecurity, between technology and manned security services, are going to continue to blur and the interaction between them will increase. The rationale is a physical threat can become a cyber threat, can become a physical threat, and so on. Having an AI-driven solution in this space is unique in that it has the potential to revolutionize the way security officers and executive staff assess risk and take proactive steps to mitigate it.
“HELIAUS is unique in the sense that it focuses on creating an ecosystem around the manned guarding security services that brings together very rich data, advanced artificial intelligence, to predict not only what's going to happen but to give recommendations on how to prevent them from happening and drive better outcomes. Then a very sophisticated and location-aware workflow engine ensures that those recommendations are acted on,” says Mullison.
“There's a saying in the artificial intelligence community that states, ‘If you have the right knowledge representation, problem solving becomes easy.’ There's a lot of truth to that. We spend a lot of time in organizing the information in our proprietary attribute model in such a way that it supports predictive analysis. It really is not just the fact that we have this very rich source of information, but it's how that knowledge is represented. Who are the security professionals on site? What's happening at this site? What sort of business are they in? What's the neighborhood like? What's the weather been like? What's the weather going to be like? All sorts of information that becomes available and organized in a way that we can reason over it and come up with understanding about what's predictive in nature.”
So Mullison acknowledges that the proprietary attribute model is the solution’s first big differentiator. The second one is the advanced artificial intelligence and representative algorithms.
“In the old world of doing any sort of predictive analysis you'd come up with a hypothesis and you'd test it against the data to see whether there was correlation, and if there was, then you’d test further to understand if there was causation. Certainly, that's super useful. People have solved a lot of problems like that over the years. But it's very manually intensive, and it requires a whole lot of focused work by people skilled to do that” continues Mullison. “In our world with HELIAUS, the AI is the one that goes through the information, reasons over the knowledge, and understands what's predictive in nature. Once it decides what's predictive, and it can do that through very deep math, then it goes one step further and prescribes what should be done in order to drive a better outcome. That's unique. But, you know, that's not enough. If you know that a stock's going to double in the next six months, that's interesting but not useful unless you act on it. You must act on that knowledge.”
It’s All About the Data
Mullison reveals this is where the third major differentiator comes in. It's this extremely robust workflow that ensures that the recommendations the AI engine is making are acted upon, are then carried out -- perhaps by the security professionals on site or by the clients themselves. It could be something as simple as a required maintenance action because of a breakdown or just regular service that a contractor can take steps to resolve.
“There's great value in each of these three components. Having a rich, robust data platform and having lots of information in one place is super powerful because you can get great situational awareness, if that information's available in real time and contains great historical awareness to help understand trends. HELIAUS has some great visualization tools and the AI is very useful because if you're too busy to notice those trends or understand what might be happening, the AI will do that for you,” Mullison adds.
“And we're super excited about the workflow engine not only because it will ensure that the AI recommendations will get carried out, but also because it is a great way to ensure that clients' post orders are carried out. Ensuring post order compliance is a challenge in our industry. HELIAUS brings another level of insurance to make sure that we are providing unparalleled post order compliance because the platform has a mobile device carried by the security professionals which becomes a virtual coach or a virtual supervisor. It will prompt the security professionals to make sure that they're doing what it is that they need to be doing throughout the day.”
The other element of the HELIAUS solution is the cloud-based Portal, which serves as a command and control center where Mullison says supervisory security personnel can employ an array of visualization tools to assess potential threats in real time and then communicate to service officers based on information from various data resources and technology end devices. He adds that this real-time communication and robust data sharing has a tangible ROI factor for the client and can cut safety and security incidents as much as 20 percent.
“HELIAUS has the ability for clients to understand their ROI in two different ways. One way is within our data visualization tool. We can enable something that we call an ROI mode which will ascribe value on the positive side or cost on the negative side to anything that happens on their site that's recorded in the system as an incident or incident avoided,” Mullison says.
About the Author:
Steve Lasky is the Editorial Director of Endeavor’s SecurityInfoWatch Security Media, which includes print publications Security Technology Executive, Security Business, Locksmith Ledger Int’l, and the world’s most trafficked security web portal SecurityInfoWatch.com. He is a 32-year veteran of the security industry and a 27-year member of ASIS.
