Answer trust model · PRD §0.2b
Unlike like-only feeds, Hoihan runs verified weight, AI spam/harm triage, expert/ops queues, and high-risk adoption gates. This is not a claim that every answer is fully lawyer-reviewed.
Verification and experience badges are ranking and report-weight signals. Non-verified members can still post.
On create, local rules + AI score spam/harm signals. Missing keys or API failure do not fail-open as fully safe.
Expert in-feed participation, official flags, and report/hold queues act as second-line ops. High-risk adoption requires source evidence or review status.
Questions structure context; answers separate official sources from experience; adoption helps the next similar user.
Trust is enforced by gates and queues—not likes alone. High-risk answers cannot be adopted without evidence, and reports feed operator review.
No. Experts can participate and ops queues handle holds/reports. There is no claim of 100% licensed full review of every answer.
No. AI only triages spam/harm risk and may assist writing. It does not issue legal verdicts.
It surfaces helpful answers for similar cases, and high-risk adoption still requires the source/review gate.
