BUTERIN: I would actually expect that, on net, it would require less institutional trust. Part of this is just because a lot of people don’t realize the sheer level of centralization in proof of work already.
In Bitcoin, there is one person, Jihan Wu, who controls the largest two mining pools in bitcoin that add out to 42 percent of the network.
And 42 percent isn’t enough to do a 51 percent attack, but it is enough to do selfish mining, which is enough to reduce other people’s profits to the point where they drop out, and you can do a 51 percent attack.
Granted, these are pools, but a very large portion of that 42 percent, I believe, actually is his own hardware.
I think, because we’re used to it, people do underestimate the oligopoly and the rich-get-richer and the trust that’s necessary inside of proof-of-work systems already.
For example, bitcoin users already trust Jihan Wu and Wang Chun to not team up and start doing 51 percent attacks pretty much every day. They seem to be nice enough or at least rationally self-interested enough not to do that, but that’s definitely still institutional trust.
https://medium.com/conversations-with-tyler/vitalik-buterin-tyler-cowen-cryptocurrency-blockchain-tech-3a2b20c12c97