Seeing the Threat: Why Counter-UAS Needs an AI Brain That Detects Aerial Threats With Vision Alone

A handful of data points illustrate the growing threat posed by drones. In the first quarter of 2025, the U.S. Federal Aviation Administration (FAA) documented 411 unlawful drone incursions near U.S. airports, and the agency notes it receives over 100 drone sightings near airports each month. In 2024, a staggering 350 drone detections were recorded across 100 U.S. military installations, according to the commander of NORAD and U.S. Northern Command.

Drones have been weaponized before - in 2020, a drone carrying copper wire was used to attempt to disrupt a Pennsylvania electrical substation, marking the first known case of a modified drone targeting U.S. energy infrastructure. Organized crime groups have also exploited drones, with Georgia authorities arresting nearly 150 individuals in an enterprise smuggling drugs, phones, and weapons into prisons via drones.

The battlefield has also witnessed the rise of drones, with small, inexpensive aircraft now causing roughly 70-80% of battlefield casualties in Ukraine, according to Western and Ukrainian officials. This conflict has served as a real-world proving ground for both drone attacks and defenses, with lessons learned directly influencing procurement plans of Western militaries.

The economics of the drone threat make it a persistent problem rather than a fleeting one. Traditional defense budgets are justified by the likelihood of an attack, but drone risk shifts the focus to the potential cost of a single incident. With a low-cost drone capable of causing significant disruptions at a fraction of the price of traditional weaponry, the attacker gains even when no physical damage occurs.

In response, counter-drone capability is evolving from a one-time purchase to a continuous, multi-layered approach akin to cybersecurity protocols.