How is AI rewriting the rules of engagement in modern warfare?
For decades, military strategists have imagined a future in which artificial intelligence determines the pace and outcome of war. That future is no longer theoretical. As tensions and conflict between Iran and Israel have escalated in recent years, AI has quietly emerged as one of the most powerful forces operating behind the scenes of modern conflicts.
The battlefield is no longer just a physical space with troops and artillery. It’s a huge invisible network of data, sensors, and machine learning models. In the current Iran-Israel conflict, AI is not just a support tool, but the central nervous system for both offensive and defensive operations. Decisions that once took hours or even days for human analysis are now made by algorithms in milliseconds. This change is clearly evident in modern air defense.
AI in layered air defense
When Iran launched an unprecedented volley of Shahid drones, cruise missiles, and ballistic missiles toward Israel, starting with airstrikes in April 2024 and continuing through the recent escalation, this sheer volume of attacks was designed to overwhelm conventional air defenses. No human operator, no matter how highly trained, can calculate the trajectories of hundreds of projectiles simultaneously in real time. This proves that conventional air defense systems are not effective enough in such intense combat situations.
This is where AI-driven defense systems become the ultimate barrier. Israel’s multi-layered defense networks, such as the Iron Dome, David Thring, Arrow 2 and Arrow 3 systems, rely heavily on machine learning algorithms to identify, classify, predict and prioritize threats, as well as optimize real-time decision-making.
Algorithmic decision-making in air defense
Detection and classification
The first step in any missile defense system is the detection of incoming objects. Radar systems continuously scan the surrounding airspace and detect objects by transmitting radio waves and receiving their reflections. Raw radar data must be processed to distinguish real threats from background noise such as birds, weather disturbances, and debris.
One of the most important tools used in radar signal processing is the Fast Fourier Transform (FFT). This algorithm converts the radar signal from the time domain to the frequency domain, allowing the system to detect movement and calculate the speed of incoming objects through the Doppler effect.
Another important algorithm is the Constant False Alarm Rate (CFAR) detection method that identifies potential targets while minimizing false alarms caused by environmental noise.
Once an object is detected, a machine learning model is used for classification. Algorithms such as convolutional neural networks (CNNs), support vector machines (SVMs), and random forest classifiers analyze features such as speed, size, and flight patterns. These models can help determine whether an object is a drone, a flock of birds, or a high-velocity ballistic missile. Accurate classification is essential to prevent unnecessary interception and use resources efficiently.
Trajectory prediction and impact evaluation
Threat detection is only the first step. The next challenge is deciding where to land. Ballistic missiles follow trajectories that are influenced by gravity, atmospheric drag, and initial rate of fire.
The system uses mathematical formulas to calculate the expected path of the projectile. To improve prediction accuracy, many missile defense systems use algorithms such as Kalman filters and Extended Kalman Filters (EKF). These algorithms continuously update their predictions based on the radar measurements they receive.
For complex scenarios where the motion of objects may be uncertain or non-linear (such as hypersonic missiles such as Fatah 1 or Kinjar), Monte Carlo-based algorithms are used. These techniques generate multiple possible trajectories and calculate the probability of collision at different locations. If the predicted impact point is in an uninhabited area, the system can decide not to intercept the projectile, thereby saving expensive interceptor missiles.
Interceptor assignment
Interceptor assignment is one of the most important decision-making functions in modern missile defense systems. When multiple threats are detected simultaneously, defense systems must quickly determine which interceptor missile or defense battery should respond to each threat. Artificial intelligence (AI) can enable this process to occur within milliseconds by analyzing threat data, evaluating possible responses, and selecting the most efficient interception strategy, as interceptor missiles are very expensive and available in limited quantities. Unnecessarily launching interceptors can reduce the system’s ability to respond to more dangerous threats later in the attack. You can evaluate multiple engagement strategies simultaneously using algorithms such as dynamic programming, heuristic optimization techniques, and reinforcement learning techniques. These algorithms consider a large number of potential engagement scenarios and select the one that minimizes the risk to the defended area.
The danger of flash wars
Integrating AI into defense systems brings significant benefits, but also creates new strategic risks. As both sides of a conflict deploy AI systems that react in milliseconds, the window for human intervention becomes increasingly narrow. There is a very real fear of “flash wars”. A scenario in which an AI radar system misidentifies a threat and automatically initiates countermeasures, setting off a chain reaction of algorithmic escalations before human leaders can even pick up the phone.
The Iran-Israel war proved that AI-driven defense systems are no longer science fiction. We handed the keys of war to algorithms and prioritized speed and efficiency above all else.
The question now is whether safety brakes can be built into machines designed to move faster than humans thought possible.
Degree of Thought is a weekly community column launched by Tetso College in partnership with The Morung Express. The Thinking degree explores the social, cultural, political and educational issues around us. The views expressed here do not reflect those of the institution. Tetso College is an NAAC accredited UGC accredited Commerce and Arts College. The editorial team includes Mr. Chubamenla, Mr. Asst. Professor of the Department of English and Associate Professor Linsit Saleo. Manager of IT, Media and Communications. For feedback or comments, please email dot@tetsocollege.org.
