AI tools pinpointed the location of the Hamas Jabaliya Battalion commander leading to an airstrike that killed him and 50 terrorists.

By Hezy Laing

In the past few years IDF has dramatically expanded its use of artificial intelligence (AI) to identify and prioritize military targets, marking a transformative shift in modern warfare.

Currently Israel’s AI targeting systems are considered the most advanced globally.

In the ongoing conflict with Hamas in Gaza, AI systems have been deployed to rapidly populate the IDF’s “target bank,” a classified list of individuals, locations, and infrastructure deemed legitimate military objectives. These include tunnel networks, command centers, and individual operatives.

In recent Israeli military operations, particularly the 2023–2025 war against Hamas in Gaza, an incredible 90% of IDF targets were generated using artificial intelligence (AI).

On the first day of the ground invasion alone, AI-assisted targeting enabled the IDF to strike 150 tunnel sites, a feat that would have taken weeks using traditional methods.

While human analysts still vet AI-generated targets before approval, the sheer volume and speed of AI recommendations have reshaped the tempo and scale of IDF operations.

One of the most prominent AI tools is “Habsora” (Hebrew for “The Gospel”), which can generate hundreds of new targets in minutes by analyzing intercepted communications, satellite imagery, and social media data.

Another system, “Lavender,” uses machine learning to cross-reference behavioral patterns and metadata to flag suspected Hamas operatives, though its accuracy has sparked debate among analysts.

Unit 8200, the IDF’s elite intelligence division, has integrated AI into facial recognition, audio analysis, and Arabic-language chatbots to locate hostages and track militant movements.

In one high-profile case, AI-enhanced audio tools helped pinpoint the location of Ibrahim Biari, commander of Hamas’s Jabaliya Battalion, leading to an airstrike that killed him and over 50 terrorists.

According to the IDF, AI does not make autonomous decisions; recommendations are vetted by human analysts before being approved by senior officers.

The IDF believes that these tools have improved accuracy and reduced collateral damage.

To build its database of targets, the IDF’s artificial intelligence systems analyze a vast array of structured and unstructured data sources, combining military intelligence with real-time digital inputs to identify and prioritize threats. This includes:

Signals Intelligence (SIGINT)
Intercepted phone calls, radio transmissions, and encrypted messages are scanned for keywords, behavioral patterns, and geolocation data. AI tools like Lavender use metadata to flag individuals based on frequency, timing, and proximity to known operatives.

Open Source Intelligence (OSINT)
Social media posts, public videos, and online forums are mined for clues about militant activity, affiliations, and locations. AI can track changes in profile pictures, hashtags, and group memberships to identify potential threats.

Imagery Intelligence (IMINT)
Satellite and drone imagery are processed using computer vision to detect tunnel entrances, weapons caches, and unusual movements. AI can compare historical images to spot new construction or camouflage.

Human Intelligence (HUMINT)
Reports from field agents, interrogations, and informants are digitized and cross-referenced with other data streams. AI helps validate or refute claims by checking consistency across sources.

Military Databases
Existing IDF intelligence archives—such as known Hamas operatives, safe houses, and logistical hubs—are integrated into the AI’s learning model to improve accuracy and reduce false positives.

Behavioral Patterns
AI systems analyze patterns like SIM card swaps, travel routes, and financial transactions. For example, frequent visits to known command centers or encrypted messaging at odd hours may trigger suspicion.

Real-Time Battlefield Data
Live feeds from surveillance drones, sensors, and troop movements are fed into the system to update target viability and risk assessments.

As of 2025, Israel’s AI targeting systems are considered among the most advanced globally.