How beTweet replaces opaque corporate moderation policy with a randomly-selected jury of verified users — and why that is the only model that scales without corruption.
Content moderation on large platforms is decided by a combination of underpaid contractors, opaque ML models, and secret policy documents. The result is inconsistent, unaccountable, and easy to manipulate by anyone who knows the rules better than the moderators do. beTweet tries a different mechanism, borrowed from the oldest accountability institution we have: the jury. When a post is reported, five verified users are selected at random. They review the post, vote to keep or remove it, and the majority decides. Jurors are not platform employees. They are not selected by us. They cannot be pre-vetted by the reporter or the reported. Severe content â CSAM, terrorism, credible threats â is escalated to admins directly and never reaches a jury. AI assists with flagging but never makes the final call. A jury has three properties that no policy document can match: 1. Unpredictability. You cannot game a moderator who does…