After spotting drones overhead in September 2025, Copenhagen Airport shut down for four hours. In the aftermath, officials canceled 109 flights and redirected 51 more. Two months later, Brussels Airport closed twice in one evening for the same reason, grounding 15 flights and diverting 8 others. In another similar drone sighting, Stockholm’s Arlanda Airport halted operations for hours. And in the United States, nearly two-thirds of near-collisions at the 30 busiest airports in 2024 involved drones, according to an Associated Press analysis.
None of this is new. London’s Gatwick Airport went through the same ordeal back in 2018, shutting down for 36 hours and stranding 140,000 passengers. What’s striking is that almost eight years later, with all the advances in artificial intelligence and sensor technology, airports and security agencies are still getting stuck in the exact same bind. While airports and security agencies can increasingly detect drones, safely stopping them remains far harder.
This gap — between spotting a threat and actually stopping it — is where counter-drone defense is stuck, and where AI’s limitations become unavoidable.
Where AI Detection Is Not Enough
AI is becoming essential to counter-drone systems. It helps with processing sensor data fast, flags unusual activity and can predict where drones are headed at scale. However, when it comes to safely stopping a drone, the limitations of AI become evident. Shooting down a drone over a crowded stadium risks falling debris that could have devastating consequences. Jamming radio signals sounds surgical until you realize it can also knock out navigation, emergency communications and even authorized drones operating nearby. In other words, the solutions available create problems that are often worse than the threat itself.
“Older legacy methods like radar for detection and kinetic interception for mitigation were built for conventional warfare, not sensitive civilian or complex homeland security environments,” said Jeffrey Starr, chief marketing officer at D-Fend Solutions. “They struggle with the core challenge of the modern drone age: distinguishing a hostile threat from a nuisance or even an authorized drone without causing chaos.”
The challenge has gotten worse as drones have become cheaper and more capable. Germany’s air navigation service recorded 192 drone-related airport disturbances in 2025, up from 141 the year before, according to Euro News. The FAA reported 200 drone sightings near U.S. airports just in the first three months of 2024, with six requiring pilots to take evasive action.
The bottom line is that detection alone does not protect runways, passengers, or critical infrastructure if there is no safe way to act on it.
The Case For Cyber Control
This is where cyber takeover offers a different approach. Instead of destroying the drone or disrupting communications, RF-cyber systems detect and understand the drone’s control signals and then, if needed and where regulations allow and operated by authorized personnel, assume command. They identify the drone’s communication protocol, stop the drone pilot’s input and guide the aircraft to a controlled landing zone. The threat is neutralized, the drone stays intact for permitted intelligence gathering and forensic analysis, and everything else continues running.
The operational advantage of that approach is continuity. At an airport, a kinetic takedown would risk debris on runways or near terminals. Jamming can interfere with communications, navigation, or emergency responders. But cyber takeover isolates the rogue drone, while allowing normal operations to continue.
“AI is a powerful tool for processing vast amounts of sensor data, but detection is only half the battle,” Starr told me. “You still need a mechanism to neutralize the threat safely. In counter-drone defense, an AI detection-only approach might tell you a drone is approaching, but if your only options are to shoot it down or blast the airwaves with noise, you have failed to solve and haven’t effectively mitigated the operational risk.”
The Cost Of Getting It Wrong
The financial and operational stakes are considerable. Kinetic interceptors can run over $125,000 per unit, according to Army budget documents. Jamming systems are less expensive, but their use in civilian airspace is heavily restricted because of interference risks. In fact, unauthorized jamming in the U.S. is a federal crime that can lead to serious financial penalties, equipment seizure and even imprisonment. RF-cyber systems operate at a fraction of kinetic costs while avoiding the operational complications of jamming.
“Cyber takeover fundamentally flips the risk equation by drastically reducing the collateral damage risk from the mitigation process,” Starr explained. “Kinetic solutions introduce the danger of falling metal and shrapnel, which is unacceptable in a crowded stadium or a busy site. Jamming is also problematic because it can cripple friendly communications alongside the target drone. In contrast, RF-cyber is non-jamming, non-disruptive and non-kinetic — it surgically seizes control of the specific rogue drone without interrupting legitimate communication or endangering bystanders.”
Still, cyber takeover is not meant to be a universal solution. The technology is most effective against mass-market, DIY and commercial drones, often classified as Group 1 and 2 systems, which account for the majority of incidents in civilian airspace and domestic security environments. It is less relevant against larger, military-grade Group 3 and above platforms or against fully autonomous drones that do not rely on common radio-frequency control links.
That distinction reinforces the need for layered defenses, a priority reflected in the Pentagon’s Replicator 2 initiative. In practice, cyber takeover functions as one option within a broader toolkit, allowing operators to address smaller, lower-cost threats without escalating to kinetic responses designed for far more capable systems.
What Procurement Priorities Reveal
The way agencies buy counter-drone systems is changing. As drone technology advances, they are moving from hardware-heavy platforms that become outdated to software-defined systems that can adapt to new threats.
“Buyers today are prioritizing adaptability and future-proofing over static hardware,” Starr said. “Drone technology constantly changes — a hardware-heavy system designed for the threats of past years is already obsolete. Agencies are looking for proven software-defined solutions like RF-cyber that can be constantly updated to counter new drone threats.”
Detection systems still dominate current spending, accounting for more than half of the $3 billion counter-drone market in 2024. But operational priorities are moving toward systems that can actually stop threats, not just spot them. That’s not surprising, especially when you consider that merely detecting a drone doesn’t protect people or infrastructure if operators lack a safe way to remove it.
That shift is also changing where decisions get made. Starr expects counter-drone response to move closer to the edge, supported by AI-driven recommendations rather than centralized command alone. “The speed of drone incursions — often appearing in swarms or at short ranges — leaves little time for a command center to micromanage every mitigation. Centralized command remains vital for the big picture, but the immediate tactical response must also be local, agile and allow for both autonomous and human decision-making.”
AI can recommend responses, but human operators usually remain responsible for final decisions, particularly in civilian settings where legal authority and public safety are paramount.
The real challenge in counter-drone defense isn’t building sharper sensors or faster interceptors. It’s closing the gap between the moment you see a threat and the moment you can safely neutralize it. AI handles the first part well. But the second part requires something AI can’t provide on its own: Surgical precision that doesn’t disrupt everything else around it.
Cyber control offers that precision, but only when it’s integrated with radar, analytics, and flexible mitigation options. Nearly eight years after Gatwick, the airports and security agencies that figure this out will be the ones that stop writing the same headlines over and over.
