Hours before dawn, under the veil of a new moon, two figures in military fatigues grapple like Greco-Roman wrestlers within the razor wire perimeter of the Blue Grass Army Depot in Richmond, Kentucky.  

Their movements are rigid but discreet, each maneuvering for leverage beneath the orange glow of the floodlights lining the depot’s security fence.

In the distance, a patrolling sentry squints, straining to make sense of the dimly lit commotion. He thumbs the two-way radio on his pistol belt but hesitates, worried he may frustrate his supervisor with an inaccurate report.  

But before the fight can go to ground — and before the sentry can reassess and request support — a bright green reticle highlights the entwined bodies on a control room monitor several miles away. Scylla, the artificial intelligence algorithm powering the depot’s security architecture, has made sense of what the sentry cannot.  

Scylla knows instantly that the image captured by the depot’s security cameras depicts a struggle. In seconds, the algorithm references a database of friendly and malicious faces, identifies the belligerents and fires a report to the watch captain: “An on-duty military police officer is trying to subdue an intruder — a known bad actor with presumed hostile intent.”  

This fictional scenario described by Drew Walter, deputy assistant secretary of defense for nuclear matters, illustrates AI’s demonstrated capabilities and underscores its potential in the realm of physical security. Last month, the Defense Department tested Scylla at the Depot. Walter described the platform as a “considerable advancement in our ability to safeguard critical assets.”  

“Its capacity to learn in real time and reduce nuisance alarms — which fatigue security personnel and inhibit responses to legitimate threats — addresses a long-standing challenge in physical security,” he said.  

DOD is looking at AI’s ability to enhance existing surveillance capabilities and threat detection, aligning with the department’s broader strategy to integrate data, analytics and AI across its operations. 

The Physical Security Enterprise and Analysis Group, which is testing Scylla, plays a pivotal role in the department’s mission to safeguard America’s strategic nuclear capabilities. “PSEAG does physical security well and has taken the reins in both the strategic and conventional realms,” Walter said. “By unifying disparate efforts under one umbrella, we can procure and field capabilities that meet Defense Department nuclear security requirements.”  

PSEAG’s interest in AI is nested in the department’s strategic priorities. Last year, in a speech on “The state of artificial intelligence AI in the Department of Defense,” Deputy Defense Secretary Kathleen Hicks emphasized the importance of integrating AI swiftly and responsibly.  

“As we’ve focused on integrating AI into our operations responsibly and at speed, our main reason for doing so has been straightforward: because it improves our decision advantage,” Hicks said. “From the standpoint of deterring and defending against aggression, AI-enabled systems can help accelerate the speed of commanders’ decisions and improve the quality and accuracy of those decisions.”  

Chris Willoughby, electronic security systems manager at the Depot and project lead for Scylla, is working to give life to Hicks’ vision. “PSEAG is testing, evaluating and training Scylla’s artificial intelligence deep neural machine learning software to detect and classify persons’ features, behavior anomalies, armed and unarmed threats and objects by evaluating video surveillance systems in real time,” he said.  

The demonstration at the Depot showcased Scylla’s ability to detect intruders, weapons and abnormal behavior using existing video surveillance systems and drones. In one instance, the software identified an armed individual climbing a water tower a mile away. “Scylla test and evaluation has demonstrated a probability of detection above 96% accuracy standards, significantly lowering … false alarm rates due to environmental phenomena,” Willoughby said.  

Walter echoed Willoughby’s enthusiasm, lauding Scylla’s unique — and cost-effective — application to existing physical security architecture. “Scylla AI leverages any suitable video feed available to monitor, learn and alert in an instant, lessening the operational burden on security personnel,” he said. “While humans still make the final decisions regarding threat response, AI augments detection capabilities.”  

Walter said Scylla’s most vital application could lie in improving the physical security of DOD’s strategic nuclear arsenal. “Scylla’s transformative potential lies in its support to PSEAG’s core mission, which is to safeguard America’s strategic nuclear capabilities when they are in the department’s care,” he said. “The ability to detect and respond to threats swiftly is paramount when dealing with assets critical to deterrence — be they Trident missile submarines, intercontinental ballistic missiles or strategic bombers.”  

Beyond the Depot, the department is exploring Scylla’s potential in cold weather and maritime environments. In the coming months, Joint Base Charleston, South Carolina, will host Navy and Marine Corps-led assessments, ensuring that the algorithm meets their service-specific demands.  

PSEAG officials said they are especially interested in testing AI’s ability to mitigate an emerging threat: unmanned systems capable of transcending domains by emerging from the sea to operate on land or in the air.  

Though it faces challenges in the form of novel threats, conditions and environments, Walter and Hicks said AI is essential to maintaining the United States’ competitive technological edge.  

The deputy secretary said the department has worked for more than a decade to be a global leader in the development and use of AI technologies. “By putting our values first and playing to our strengths — the greatest of which is our people — we’ve taken a responsible approach to [artificial intelligence] that will ensure America continues to come out ahead.”  

PSEAG’s investment in AI is born from the 2022 National Defense Strategy‘s commitment to “driving commercialization … in emerging technologies,” like “artificial intelligence and autonomy,” and the algorithm’s noteworthy performance in Richmond, Kentucky, marks a fundamental milestone on the department’s path toward improved physical security and comprehensive AI adoption.  

“By embracing advanced technologies, like Scylla, we are not just enhancing our current security measures,” Walter said, “we are setting the foundation for future innovations that will keep our nation safe.”

This Defense news article "Defense Department Tests AI Software, Advances to Improve Physical Security Posture" was originally found on https://www.defense.gov/News/News-Stories/