This does not scale, the amount of abuse is huuuuge. But I think with a prerequisite, it could:
Companies should be required to provide access to a service that verifies identity. I know such companies exist, so it is doable. And then, once it is provable that they are dealing with an actual human who can be identified, your rules can be applied.
Apple made 100 billion profit last year. They can surely afford to make this. Just because it would cost them profit does not mean we shouldn't require it.
For Apple, yes, but in the context of rules that apply across the board we should address the scaling issue. People who've had to deal with the filth of the Internet know how hard the problem is to solve, and not everyone has Apple money.
If you can't charge your customers enough to spend enough on this challenge, you don't really have a viable business, you've got a theft organization. Externalizing your failure to build a solid business by screwing customers is not okay.