Since people are talking about more moderation controls again I want to point out the method I am most excited about that I think would have the biggest impact:
Giving users control over how their replies are handled and dissiminated in ActivityPub.
This would require a change to the protocol such as FEP-5624 https://codeberg.org/fediverse/fep/src/branch/main/fep/5624/fep-5624.md
I would also like to see a cryptographic step that requires Instance C to validate a reply from Instance B based off of a cryptographic signing on Instance A. This would allow granting of access to replies and revocation of replies.
Of course, malicious actors can write code to post replies on their own instance without permission just as they can start their own blog and screenshot a post and write a response on there. However, if the protocol change is adopted, then it would significantly inhibit the spread of replies which do not have consent to respond.