12.11.24 – SSI – Peter Giacalone
Almost all mainstream applications of AI technology today (e.g., ChatGPT) are narrow AI, no matter how much they might look like general AI.
Over this past year, through my articles, seminars and social media posts, I have sought to evangelize detailed messaging. These articles, seminars and social media posts detailed the significance and essence of the following technology subjects:
- remote video monitoring and guarding
- artificial intelligence (AI) presence detection as part of intrusion, life-safety and aging-in-place systems
- conversational AI for improved service levels and customer satisfaction within monitoring centers and full-service operations
Artificial intelligence, more commonly known as AI, has to be one of the most, if not the most, overstated phrases of 2024. However, as I will briefly outline, AI is real and, when leveraged responsibly by credible platforms, delivers tremendous value.
I say AI is overstated because people love throwing the phrase around and, in so doing, tie it to many things that are simply not AI. That makes it a bit confusing for some who have a genuine desire to understand artificial intelligence.
Two Main Types of AI
At the highest level, there are two main types of AI: “narrow AI” (an AI that is trained to solve one specific problem as accurately as possible) and “general AI” (an AI that can solve any generic problem and can think critically and reason about any topic like a human would).
Almost all mainstream applications of AI today (e.g., ChatGPT) are narrow AI, no matter how much they might look like general AI.
ChatGPT is in the class of “generative AI,” which is trained on an input. Although ChatGPT appears to be a general AI, it lacks the generic reasoning necessary to garner that distinction; instead, it is most accurately classified as narrow AI.
As of today, a true general AI has yet to be created. However, we have been able to solve a lot of real-world problems with narrow AI.
This is important to truly understand how AI relates to all three subjects I have been heavily communicating over the past year.
It’s important to understand that ChatGPT leverages all text provided by humans that resides on the internet and is trained to do very specific tasks — namely, respond to the text input and provide a text output that is a relevant answer to that text input.
Similarly, the AI leveraged in the subjects I have communicated is similar, serving as a robust ecosystem and delivery mechanism.
What follows are some details:
Remote Video Monitoring and Guarding
Remote guarding and video monitoring are supported through specialized and traditional monitoring automation platforms. However, a variety of AI or video analytic technologies are embedded to monitor and measure a variety of actions, habits, movements and scenarios.
When they match what is sought out, they deliver notifications, alerts and whatever is predetermined for further analysis.
AI Presence Detection Technology
Presence detection is provided by detecting a variety of elements, depending on the platform that is utilized. Wi-Fi sensing monitors for movement, lack of movement, movement types and the absence of movement.
It compares this to many models and historical data as a means of delivering results and notifications.
To better understand the specifics of physical presence, others utilize the monitoring of wireless devices. By combining state-of-the-art data streaming capabilities and cutting-edge technology with advanced AI and machine learning algorithms, it’s possible to deliver presence insights that incorporate actual specifics of who may occupy a premises real time.
Conversational AI Technology
Taking the leveraging of data and behaviors, as previously described, and integrating it with natural language speech takes things to a new level. The older IVR technology also leveraged natural language speech, but it relied on pre-programmed scripts to understand and respond to user queries within the monitoring automation systems.
Conversational AI with natural language speech, by contrast, presents great progress. Powered by machine learning and natural language processing algorithms, conversational AI systems possess the ability to understand, interpret and respond to user inquiries in a more dynamic and contextually relevant manner.
That’s quite different from a siloed approach in which interactions must meet the structure and format of the IVR.
As you might surmise, leveraging AI for detection, monitoring and supporting the security and life-safety industries is beneficial. The common thread is better customer service, precise and immediate detection, and the ability to bring context to the services we already provide. This elevates the all-around value proposition for the end user.
About the Author
Peter Giacalone
Peter Giacalone is President of Giacalone Associates, an independent security consulting firm.