I agree that there is no universally accepted general definition of intelligence , there are some very specific subset that can be somewhat reliably measured but that’s not even close to a general score.

Standardised tests measure the ability to pass the tests, last i checked it was mostly rote memorization more than applications of skills learned. ( my info could be out of date or incorrect on this )

When scores are lower is doesn’t mean the students are dumber. It means the school system is failing to teach.

That’s not strictly true and it’s also not a mutually exclusive scenario.

For a given metric of “dumbness” it is possible for it to go up and down independent of school teaching standards and approaches. or even a combination of the teachings standards + some other things

Also if you are going to argue that perceived “dumbness” can only be attributed to failing to teach , that assumes a usable metric for “dumbness” that can be measured.

I’m mostly agreeing with your points, i’m just not seeing how “can’t measure smartness” and “smartness can only be falling due to teaching failures” can be posited simultaneously.