Hive is yet to prove long term sustainability of its model. Especially if it takes off to use levels comparable with other social media out there. It is good that Hive can be developed without rush in steady pace, I believe in some aspects we are already "overdeveloped", that is, prepared for much more traffic than we currently experience and/or allow (cough, cough, max block size). However important parts of second layer, Hivemind in particular, still need work.
As for Twitter. With the global enterprises like that, there is always a problem: what is cheap in one place is prohibitive in other. If the proposition is to have dual system - you pay, you are the customer, you get extra features and freedom of speech, or you don't pay and therefore you are the product, that is going to keep being moderated and molded for Twitter benefit. It might work. Although if it turns out being cut along geographical lines, it might be better to make two/more separate Twitters that don't interact with each other - one paid only, for developed world and the free one for the rest. Bots would not be as common in paid version due to fees, and would avoid unpaid version due to lack of incentives to run there.