The Tragedy of AI Warfare: Ghosts of Gaza

What do we lose when algorithms decide targets and bombings are approved in just 20 seconds?

• How Israel’s AI-based targeting systems ‘Lavender’ and ‘The Gospel’ operate
• How the common belief that AI warfare reduces civilian casualties was shattered
• The impact of the Gaza crisis on future warfare and international norms

An officer from Israel’s elite intelligence unit stares at a monitor. The man on the screen is not a target chosen through human deliberation. His fate was decided by an AI warfare algorithm, and the officer approved the strike in just 20 seconds. This article delves deeply into how artificial intelligence created the tragedy known as the “mass assassination factory” in Gaza, exploring the system’s operation and its horrific consequences. The trauma of war ironically became a ‘moral disengagement device’ distancing humans from the consequences of violence.

Architects of Destruction: The Reality of AI Warfare Systems

At the heart of Gaza’s tragedy are multiple AI systems working organically. These are not mere support tools but the core engines amplifying the speed and scale of killing to unprecedented levels.

Lavender, the Human Hunter

‘Lavender’ is an AI database developed by Israel’s Unit 8200 that analyzes vast surveillance data on Gaza’s 2.3 million residents to identify suspected militants. Early in the war, Lavender generated a “kill list” naming as many as 37,000 Palestinian men as potential targets.

The system assigns individuals a risk score from 1 to 100. Even ordinary behaviors—such as belonging to the same WhatsApp group as suspects or frequently changing phone numbers—could trigger suspicion. In other words, the everyday life of Palestinians was criminalized by the algorithm.

The Israeli military approved the system’s use despite knowing its error rate was as high as 10%. This means 1 in 10 targets could be innocent, and this was not a flaw but an intentional design feature.

The Gospel (Habsoora), the Building Hunter

While Lavender targets people, ‘The Gospel’ targets buildings. This system accelerated target generation to an industrial scale, creating what a former officer called a “mass assassination factory.” Where human analysts might create 50 targets a year, The Gospel generated 100 new targets daily.

It identified not only military facilities but also civilian high-rises, universities, and other structures as “power targets,” a concept with no basis in international law.

“Where’s Daddy?”, the Final Piece in the Killing Process

A location-tracking system named “Where’s Daddy?” captures the moment a Lavender-identified target enters their home, triggering an airstrike. An intelligence officer testified, “It’s much easier to bomb a family’s home.” This reveals a deliberate tactic to maximize civilian casualties.

Speed Swallows Judgment: The New Doctrine of AI Warfare

In Gaza, AI was used not to protect civilians through precision strikes but to maximize the speed and scale of destruction. An IDF spokesperson admitted, “The focus is on damage rather than accuracy.”

The system continuously supplied targets, and one officer lamented, “There are another 36,000 targets waiting for you,” revealing the immense pressure. AI was not a precise scalpel but the engine accelerating a ‘mass assassination factory.’

The Acceptable Calculus of Death

The IDF set pre-allocated quotas for permissible civilian deaths per airstrike:

• Up to 15–20 civilians could be killed to eliminate one low-level militant.
• Killing a senior Hamas commander could justify over 100 civilian deaths.

Cheap unguided bombs, or “dumb bombs,” were primarily used. One officer explained, “We don’t want to waste expensive bombs on unimportant people.” This chilling testimony revealed how human life was ranked and disregarded under the guise of efficiency—one of the most shocking findings of this investigation.

“My Children’s Small Bodies Were Torn Apart”

Behind the statistics lies human suffering. The ‘output’ of the AI factory is not data but sacrificed lives.

Al Jazeera journalist Wael Dahdouh received live news that his wife, son, daughter, and grandson were killed by an Israeli airstrike while seeking refuge in a camp. His tragedy highlighted to the world that there is no safe place in Gaza.

Writer Ahmed Alnawk lost 21 family members—including his father, siblings, and nephews—in a single strike. According to an Amnesty International report, Islam Harb lost all family members except his four-year-old daughter Lin in an airstrike and testified, “My children’s small bodies were torn apart.”

Can this truly be called a ‘precise’ AI war? The more data-driven the war becomes, the more victims are dehumanized—a tragic paradox.

Gaza, the Laboratory of AI Warfare

Gaza became what the Israeli military called the “first AI war” laboratory. The entire Palestinian population was reduced to a ‘data body’ for AI training through facial recognition, call interception, and social media surveillance.

More disturbingly, data from airstrikes is used to improve the system. The deaths of Palestinians become R&D resources for future wars. What happens in Gaza sets a technical and ethical precedent for future global conflicts.

Global Impact and the “Killer Robots” Debate

The Gaza crisis ignited international debate over Lethal Autonomous Weapons Systems (LAWS). The UN Secretary-General and the International Committee of the Red Cross (ICRC) urgently called for treaties regulating LAWS, emphasizing that “autonomous human targeting by machines is a moral line we must not cross.”

However, major powers like the US and China are heavily investing in the AI arms race. The technologies tested in Gaza are only the beginning of a larger global competition.

Conclusion

The AI war in Gaza poses heavy questions. When a system with a 10% error rate causes a family’s death, who is responsible—the programmer, the commander, or the machine itself?

• AI has changed the nature of war: It was not a tool to increase precision and reduce harm but an engine accelerating a ‘mass assassination factory’ maximizing killing speed and scale.
• Human moral responsibility vanished: Procedures like ‘20-second approval’ excluded human ethical judgment and encouraged ‘moral disengagement,’ hiding responsibility behind machines.
• This is a warning to the world: Gaza is a grim preview of a future where life-and-death decisions are delegated to algorithms, underscoring the urgent need for international regulation of autonomous lethal weapons.

I hope this article raises awareness of the tragic reality and ethical issues of AI warfare. Please share this information to start a societal discussion on how technology should be used for humanity’s benefit.