Two questions that produce similar-looking outcomes
Authorization asks whether an actor was permitted to take a given action, evaluated before or at the moment the action happens. Content moderation asks whether something already posted is appropriate to keep visible, evaluated against standards about content itself, often well after the fact. These are genuinely different questions, but they get conflated because both can result in the same visible outcome — something blocked or removed — which makes it easy to treat one as a substitute for the other.
A fully permitted actor, correctly authorized to post, can still produce content that moderation later determines is inappropriate. A properly functioning authorization system and a properly functioning moderation system are both independently necessary; strengthening one doesn't compensate for weakness in the other, because they're catching different kinds of problems.
Where the conflation causes real gaps
Treating moderation as if it also handles authorization leaves the authorization question effectively unanswered — an unauthorized actor whose specific content happens to be unobjectionable sails through a moderation-only check that was never actually evaluating whether they should have been allowed to act at all. The reverse gap is just as real: an authorization check confirming the actor was permitted says nothing about whether what they posted is appropriate. Each system needs to actually do its own job.