An interesting perspective. Not entirely sure that's the same line of reasoning advertisers use - they may be more risk-adverse and just go for a blanket "we don't want anything to do with <objectionable thing>" than "we don't want to show up next to <objectionable thing> unless the user likes it"
Yeah I don't think advertisers want their ads shown if the user likes the bad content. It is just way less severe than if it was happening for all users. Also harder for Twitter to notice and fix if organic occurrence is extremely rare.
