Because it’s feedback on how effective their targeting has been when confronted with whatever electronic warfare and misdirection Israel was using to defend themselves.

That sort of information might let the attacker make adjustments to be more accurate next time

Not exactly. There are some species which haven’t changed all that much for millions of years, and those have certainly managed excellent adaptability.

Others, though, might find themselves evolving to cope with the climate right now at the expense of being vulnerable to some future problem. Say the climate is very hot, but in a few tens of thousands of years there’ll be an ice age. An animal which is well adapted to the ice age will probably go extinct before it arrives, having all been eaten by an animal well adjusted to the heat which is here right now.

“In the end” isn’t useful if you get outcompeted in the meantime

People keep talking about this work as if it offers some meaningful insight into how these systems behave, when the whole premise requires a fundamental misunderstanding of what GPT is actually doing.

The idea that you could present it with information but ask it not to use that information is absurd. At the end of the day it’s a sophisticated word association engine. It doesn’t have intent, it isn’t capable of strategy.

This is like pointing out that a dog which has learned to shake hands is actually deceiving you, and it doesn’t really mean any of the social things we use handshakes for (except it’s worse, because the dog is actually capable of being sociable)