"Trustless" is relative. In this case, the protocols are still trustless—the trust lies one meta level up, where you decide which protocols to use. In this situation, there are network effects as well—you can come up with your own perfect protocol in a vacuum, but if nobody uses it, who cares?
In this case, you can think about it as one person saying "I think it's a good idea if we do this" and a few hundred other people saying "I agree with you". Obviously, they're more likely to agree with the first person if they have some sort of institutional legitimacy. But is that "trust"? That's very different from the definition of "trust" that's generally used in protocol design.
