You can evaluate the claims much more easily. You don't have to be a genius, one can ask some basic questions and proceed from there. Sound reasoning is in short supply in this area, in any case.

• look at the code, it's open source. is it well documented, does it make sense, what do the changes look like, etc.
• read the references papers, etc.
• look at occurring transactions, number of "dapps", etc.

As far as I know there are no dapps in operation (perhaps some can correct with some actual data). Vitalik recently tweeted an example of the best use cases, and they are all no in production. I have yet to see someone demonstrate an actual practical use se, live in code. All the tutorials I've seen where about hypotheticals. And if you look at talks where they talk about the worlds operating system, it's the same. It is all theory. For example the most basic use case is registering a name. The DNS example was used always as the most simple. Yet it is most unlikely ethereum can replace DNS and its not used. The same with the claims about "web3", "IoT",... all claims about some magical future.

FWIW I received the whitepaper in December 2013 I read the paper for 10 minutes and concluded it will not work, based on the fundamental datastructure of a tree (GHOST trees). I've looked at the code but can't figure out what it does, and didn't spend much time on it. On the upside the project familiarised a lot of people about the concept of smart contracts. It's hard for me to see where the 400M$ valuation should come from, but as I remember Litecoin peaked at 500M$ so its not all that unusual.

I think it is much more productive to think things through without constant appeal to science fiction. What could these systems actually do, and how could they be useful? Then in terms of implementation that's more complex, but one hint here. Bitcoin implements a partial order of transactions and SC's need total order. It is not that hard to come up with some models of event ordering.

This is an interesting post, and shows more the direction, although I don't agree with NS on ethereum.
unenumerated.blogspot.com/2014/12/the-dawn-of-trustworthy-computing.html