inbound - 1:17 pm on Apr 11, 2012 (gmt 0)
I'd say it should be easy for Google to spot the blatant cheats; they know a great deal about the visitor levels to sites and where people go (so sites with little traffic, unless very niche, are not going to get 2356 reviews for a single item, also they could find out if toolbar/chrome users are actually using any review system). That does not mean that Google is currently using that data or even collecting it in that particular way.
I'd steer clear of doing anything unless you really do have the data to back it up.
I think an interesting area to look at is where "reviews" pop up for things that are hard to get reviews for (or even for things that do not have legitimate reviews at all - entire sectors can fall into this definition). If companies such as Google, Yelp and many others cannot get enough people interested in reviewing some verticals then you have to wonder if a goldmine of "reviews" on a small site is genuine (it's either a really good genuine site or it's one that should be dumped from the SERPs).
As for reviews being hard to get, that's so true that recommendations are pretty much the only way to get enough data in many areas. Add into that equation the need for search engines to know who a person is to be able to trust a review/recommendation (statistically speaking) and you can see why Facebook has another ace card that is scaring Google to the extent that they are going all-in on Google+.
I think we are seeing moves by Google that are motivated by the existential threat that Facebook poses, and so far Google has not struck upon a winning strategy (they, or someone else that they buy, may find a way to extinguish or reduce that threat). A big problem for Google is that advertising revenue and social recommendation features do not sit very well together; both Google and Facebook are probably working on how you monetise a system that is driven by public opinion.