Look at YouTube and the multiple adpocalypses as an example. They have a massive bank roll and engineering staff but still fall foul of journos writing a story

Look at YouTube and the multiple adpocalypses as an example. They have a massive bank roll and engineering staff but still fall foul of journos writing a story about "look at this content next to X's advert. We reached out to X for their take." which leads to a tanking of CPM, creators getting demonitised over news reporting, content creators self censoring to the determent of those with visiabilty issues (or simply being on mobile with a small screen) as instead of reading out a statement they will display it on screen in fear of YT's bots flagging because of their spoken content. No content moderation system will be 100% perfect, esp when you take it as a worldwide problem. And (atleast in the western world) facing demands for more and more content moderation. In an ideal world we should be able to leave it up to the user. But Twitter has self filters for years and centeral filtering on the platform has only increased not decreased.