As exciting as recent advancements in AI are, it’s important to avoid letting the hype train overrun the tracks. Instead, it’s important for organizations to understand where and how AI is already being used, and to explore where it may be able to add the most value to the business, examines Fredrik Nilsson of Axis Communications.
Since the start of this year, generative artificial intelligence (AI) tools like ChatGPT have taken the world by storm. It’s understandable – while AI-based solutions have long been used in a wide variety of industries, those solutions often operate behind the scenes. On the other hand, advancements in generative AI have been highly publicized – and members of the general public suddenly realize just how far artificial intelligence has come. Capabilities, once viewed as the stuff of science fiction, are fast becoming part of everyday life.
That said, it’s important for organizations to maintain a realistic understanding of AI’s actual capabilities—and to be wary of any provider promising them the world. AI can be an extremely useful tool, particularly in areas like security, business intelligence, and operational efficiency—but generative AI isn’t going to be a one-size-fits-all solution.
AI Is Making Today’s Security Solutions Smarter
Within the security world, AI-based solutions have been helping organizations keep their locations safe for quite some time. Many people still have an image in their heads of the lone security guard keeping an eye on dozens of wall monitors, but the truth is that this perception hasn’t matched reality for some time. Analog cameras have, by and large, been phased out in favor of modern internet protocol (IP) cameras, connected to a network and capable of integrating with audio solutions, access control systems, and other security solutions.
Perhaps more importantly, the cameras themselves have grown more powerful. This doesn’t just mean higher quality images (though image quality has improved dramatically), but significantly more processing power. In the past, analytics could not be run natively—video feeds would need to be sent to a central server (often in the cloud) to be processed and analyzed. Unfortunately, computing power, bandwidth, and storage capacity quickly became limiting factors for many organizations. Thanks to modern chipsets, today’s devices are capable of running powerful, AI-based analytics at the network edge, sending only the relevant metadata (and perhaps short, image-tagged video clips) to the cloud. The upshot is that modern analytics are accessible to just about any business using a single surveillance device.
It’s hard to quantify just how much these AI-based analytics have improved modern security capabilities. Today’s cameras can be programmed to alert on suspicious activity in real-time, notifying security personnel if a trespasser is detected in an off-limits area, if an individual or group of individuals has been loitering near an entrance, or if someone is engaging in aggressive or dangerous behavior. Integrated with audio solutions, they can instantly recognize sounds like cries of pain, breaking glass, and even gunshots.
They can tell the difference between a human figure and a deer, even in low-light environments. Capabilities like object recognition and tracking have become significantly more accurate and advanced as AI has grown more powerful, and they are helping organizations turn surveillance devices from reactive tools into proactive ones.
See More: MLOps Best Practices: Optimizing AI with Quality Data
AI Is Already a Business Enabler
Although many of these AI-based analytics have their roots in security, organizations are increasingly identifying ways to leverage them in ways that advance the business as well. The devices monitoring suspicious activity can also be equipped with more business-focused analytics. For example, a retailer might be interested in analytics capable of counting how many customers are in a store at a given time or tracking the movement of customers throughout the store. Or they might be interested in tracking what percentage of customers arrive on foot or how many are taking advantage of a certain promotion.
All of this information can have a real impact on a retailer’s bottom line. It might allow them to allocate staff more effectively around their busiest hours or change the store layout to facilitate traffic more smoothly. It can also tell them which products are drawing the most attention so they can be placed in strategic locations and help them better understand which marketing campaigns have the greatest impact. Simply notifying staff when a queue is forming can enhance the customer experience by dramatically reducing wait times.
Retailers are an easy example, but the same principle applies to a wide range of industries. Manufacturers can look for ways to improve movement on the factory floor or look for defective products on assembly lines. Energy companies can monitor remote locations for potential maintenance issues, helping to avoid costly downtime. Cities and municipalities can track traffic patterns, accident data, and crime statistics to deploy emergency and law enforcement personnel more effectively. These insights have a real, measurable impact on efficiency and operating costs, helping organizations better manage their expenses and streamline operations—and those who fail to recognize the value of AI are effectively leaving money on the table.
Of course, it’s important to touch on the potential of generative AI. While the real-world capabilities of the technology remain largely unproven, its potential is certainly intriguing—from both a security perspective and a business one. Some speculate that generative AI solutions may improve reporting functions, providing additional context to alerts and giving security personnel a complete understanding of what is happening on the ground. They may also be able to identify and report on patterns in shopper behavior that human observers would miss, adding new business intelligence insights to capitalize on. In short, the hype cycle for generative AI is still early, but businesses should begin looking for ways to leverage it as it becomes more advanced and accessible.
Embracing AI with Cautious Optimism
Generative AI solutions have captured the public’s imagination—with good reason. As these solutions become more advanced, they may change the world in ways we cannot yet anticipate. But artificial intelligence is not a new technology—in fact, organizations may not even realize that they have already been leveraging AI in their security solutions for years. Before moving on to the “next big thing,” organizations should ensure they are getting the most out of the AI solutions they already have. Today’s technology is already capable of adding significant value to the business—and those not taking advantage risk being left behind.
How are you enabling business with AI? What are your primary challenges? Share with us on Facebook, Twitter, and LinkedIn. We’d love to hear from you!
Image Source: Shutterstock
