re 2: yes, and that is also something I think that can help us in the “battle” against AI content. On an individual content level it’s already near impossible to tell if content is created by AI or human, that battle seems lost.
But, if we take that detection to an account level… It’s going to be much more expensive for bot networks to setup and maintain believable long-lived accounts that shows cross-app behaviors. Not saying it’s impossible, but if it’s more expensive that it’s worth… than this could help.
re 3: for Sifa specifically, I have no plans to verify accounts. What I plan to verify is claims in the account. Like the claim that you finished a course. The claim that you published a scientific paper. The claim that you work somewhere.
And even within these, there might be several ways or levels of doing this: your work could be verified by yourself through work e-mail/domain/SSO, by colleagues, by the employer themselves etc etc.
That you are who you say you are is ok, but there are many scenarios where that is not relevant. Say I’m looking to add contributors to my open source project. If your account has done a lot of great open source contributions (and this is verified), that is much more valuable and telling about you than if your name is exactly the name in your passport. I couldn’t care less 
. But for someone else in other scenarios, your open source contributions might be totally irrelevant. Context matters, a single verification icon won’t solve you being verified in a single context or giving you a single score (what Klout once did/tried).
Well we can’t dictate what appviews do, but that is our proposal yes. It’s point 3 “**Issuer always named.”
re 4: I don’t know if this will protect minors, that’s not the goal (but if that’s the positive by-affect I’ll take it). I also don’t believe ID verification is the answer to “how do we protect minors online”: It’s not going to help much, and it has a lot of other negative consequences (see for example Why age verification misses the mark and puts everyone at risk - European Digital Rights (EDRi) ).