AWS has a terrible reputation as being the source of absolute massive amounts of abuse and poorly written scrappers and crawler (hard to tell the difference between a bad crawler and an active attack).

From experience I've seen AWS be the source of overwhelming traffic so many times that we in some cases resorted to the same solution, blocking AWS completely.

I don't know if AWS doesn't care or is just slow to react. Maybe reporting is to difficult, I don't know.

So the obvious you're missing is: AWS IS a huge source of "bad" traffic and getting a misbehaving customer shutdown is too hard, while renting insane amounts of capacity is too easy for bad actors.

Edit: I've almost never seen GCP or Azure being the source of the same amount of crazy traffic.