But bots can't walk into most businesses in person. My point is that I don't want to hear from an account that isn't at least verified to be set up by a verified specific human being.

And you should be able to filter to only accounts that are verified and pay $5/month. I think this is an issue with twitter and their product. "verification" is one possible solution. But, other product changes would probably work better. E.g., I rarely have such negative interactions on hacker news and they don't even require a CC or a phone.

Agreed, and I want to see more than one change and some experimentation, too.

But if you're out in public and you start shouting harmful, hateful or abusive messages there can be real-world consequences.

On Twitter you can be the most vile person possible with absolutely no fear as you're entirely anonymous.