„Die Maschine hat es eiskalt gemacht“: Israel nutzte KI, um 37.000 Hamas-Ziele zu identifizieren

The Israeli military’s bombing campaign in [Gaza](https://www.theguardian.com/world/gaza) used a previously undisclosed AI-powered database that at one stage identified 37,000 potential targets based on their apparent links to Hamas, according to intelligence sources involved in the war.

Israel’s use of powerful AI systems in its war on Hamas has entered uncharted territory for advanced warfare, raising a host of legal and moral questions, and transforming the relationship between military personnel and machines.

“This is unparalleled, in my memory,” said one intelligence officer who used Lavender, adding that they had more faith in a “statistical mechanism” than a grieving soldier. “Everyone there, including me, lost people on October 7. The machine did it coldly. And that made it easier.“

Another Lavender user questioned whether humans’ role in the selection process was meaningful. “I would invest 20 seconds for each target at this stage, and do dozens of them every day. I had zero added-value as a human, apart from being a stamp of approval. It saved a lot of time.”

Insanely dystopian and dangerous.

By using AI to make these life and death decisions they’re systematically reducing individuals (REAL men women and children all of whom go just as deep as you or I) to numbers.

Stripping the human element from war can only serve to dehumanize it and the institutions engaging in it

Pretty convenient how you can’t throw a computer in the Hague.

Hey guys, the main plot of *Captain America: The Winter Soldier* was not supposed to be inspiration.

damn I didn’t know charity workers worked for Hamas

Seeing how inaccurate and shoddy AI is still, you can bet a whole bunch of innocent people were murdered with the help of these programs. A real bleak dystopia we are in

I mean, yeah, but a modern C4ISR system with a target selection feature that makes its choices on a Hi-Lo algorithm from the Minsky era is AI, and those are a decade old. C3I systems with stochastic search and identification features are three decades old and arguably more advanced. In warfare, it’s not so much the algorithm as it is the verification method. The Aegis system that shot down an Iraqi passenger jet full of innocent civilians wasn’t caused by its algorithms as much as it had a bad information source, a phantom track, and an overzealous commander who didn’t want to verify before he pulled the trigger. Arguably, AI would be less retributive in its motivations than a human target selector, but if its fed bad data, the input parameters and tolerances are overly permissive, or its outputs haven’t been appropriately validated before selecting a target, then its arguably no better and no worse than a person in the fog of war.

The problem with this article is that there isn’t enough information to make a judgment and it seems the leadership over relies on the intelligence this tool provides without validating its results. We can’t judge from the facts provided whether its garbage-in-garbage-out in which case why are you using AI or that nobody validated its logic in target selection at any time. I mean, its made some really bad calls.

The other side of this is economic. Russia has pioneered what some have called „sustainable warfare“ or „industrialized warfare“ which has been unopposed up until the Ukranian conflict. They’ve done that by combining old school Russian warfare tactics with a wasteful but cheap way of conducting warfare, one that pushes the limits of the Geneva conventions to the breaking point; collateral damage being just a cost of doing war that can be ignored. Israel seems to have taken the same approach here in taking the most cheapest route to eliminate targets, dumb bombs dropped on a militant’s home, regardless of who is in the building.

… I’ve seen a codebase from government defense contractor. Not sure if this thing is even unit tested, let alone moral to use to kill people. Even if it is morally okay to use (which it is not), I imagine this thing is horribly unreliable. So only thing it removes is a person’s guilt in pulling a trigger.

Next stop, „A Taste of Armageddon“ (Star Trek TOS S1e23) You were killed in action.

Step into the disintegration chamber.

Wait until people hear how they ID’d targets prior to this…

Correction: Israel used AI to JUSTIFY attacking 37,000 places by claiming it was Hamas related. It’s quite clear that Israel’s targets are schools, hospitals, aid and relief agencies, journalists, and after that women and children. Any Hamas fighters who get killed are collatoral damage.

if target !=“member of IDF“

{

order=shoot to kill

}

AI my ass 😀

And Biden said china AI endeavors is dangerous because they will use it for military use

Whose using one now, pfft US allies ofc