In a word: While some view AI as a threat, the military sees it as a tool to detect threats. It recently deployed an AI surveillance system that can identify threats such as armed individuals from more than a mile away. The system also generates fewer false alarms than those generated by trained security professionals.

Getting started with security Scylla offers “proactive” AI-based security systems to protect perimeters around facilities and depots. Its Scylla AI systems are apparently good enough to protect US nuclear sites, as the Department of Defense (DoD) has begun. essay eight months ago at the Blue Grass Army Depot (BGAD) in Richmond, Kentucky.

Currently, BGAD is the only military base testing AI-based surveillance algorithms for detect potential threats. The systems help staff find and identify intruders, weapons or “abnormal behavior” in real time. Scylla systems work with existing surveillance cameras and drones to monitor facilities, providing significant improvements in the effectiveness of human personnel threat response.

Scylla AI learns in real time, reducing false alarms, according to Drew Walter, assistant secretary of defense for nuclear issues. The system addresses one of DoD’s long-standing physical security challenges: improving security personnel reaction times while filtering out non-security issues quickly and reliably.

In BGAD tests conducted by the Physical Security Enterprise and Analysis Group (PSEAG), Scylla showed that it could detect threats with accuracy rates greater than 96%. Chris Willoughby, director of electronic security systems at the depot, said the system significantly reduced false alarms caused by “environmental” phenomena. Humans must always decide whether to respond to a threat.

The AI ​​showed remarkable surveillance capabilities by identifying an armed individual climbing a water tower a mile away. Another example of the system’s reliability is an alert sent to security personnel “within seconds” after the algorithm detected two potential armed intruders crossing a fence. The intruders were BGAD personnel and Scylla immediately identified them using facial recognition.

While PSEAG is heavily involved in testing, evaluating and even training Scylla’s deep learning algorithms for BGAD, the Army has not provided any specific details on how the trained system differs from commercial software for obvious reasons. Deputy Assistant Secretary Walter is a fan of AI because it could be “transformative” for PSEAG’s core mission: safeguarding America’s strategic nuclear arsenal.