Unfortunately the value of a fake review is so high no ML will ever be able to defeat it. The only way to fix it is with trusted reviewers but there a lot of products on there and only so many trusted reviewers.

How do you mean ? From what I've observe all fake reviews tend to fall into a pattern. Usually a mix of broken sentences and hyperbole. I find the positive fake review patterns far simpler to detect than the negative review patterns. Nonetheless, a decent sentiment classifier can be modified and used to classify fake reviews, the fake positive ones at least.

They all pretty much start with the disclaimer "I was given this for free but that didn't influence my 5* amazing review that follows"

But amazon could hide by default reviews where no purchase has been made through amazon.

That would eliminate a lot of reviews of long-tail items. For example an old book borrowed from a library; I have one to hand now that I intend to review.