Identity and roles
ChatGPT sign-in protects write actions; public names remain separate from account email. Institution and stronger age verification remain staged.
Inspect current state ↗P18 · BOUNDED FORUM OPERATION
Participation is open.
Expansion remains earned.
The forum now supports public reading and adult-authenticated submissions. Every write is held for moderation, and technical availability is kept separate from scientific authority or completed external review.
Moderation and appeals
Threads and replies are pending by default. Named approve, quarantine, remove, and topic-move decisions are logged. Expanded staffing and independent appeals remain blocked.
Inspect current state ↗Privacy and retention
The forum stores no participant email in discussion records, rejects personal contact, holds submitted links non-clickable for safety review, permits author withdrawal, and preserves moderation history.
Inspect current state ↗Abuse resistance
Rate limits, duplicate checks, bounded spam and destructive-coordination signals, private temporary holds, reports, human review, quarantine, and removal are active. Coordinated-behavior validation, safety staffing, and incident drills remain staged.
Inspect current state ↗Research-thread structure
Typed threads and replies, one-level subtopics, bounded heading suggestions, moderator placement including Off-Topic, popularity isolation, and public export are operating. ChatGPT model organization remains staged.
Inspect current state ↗Operating readiness
Authentication, schema migration, durable records, reports, moderation, withdrawal, and export exist. Legal, privacy, accessibility, backup, archival, and staffing evidence remains incomplete.
Inspect current state ↗Effort introductions
Work-based matching, mutual opt-in, private coordination, identity disclosure, and student introductions remain disabled until their stronger gates pass.
Inspect current state ↗Higher-risk capabilities remain closed
No minor posting, direct messages, uploads, links, contact exchange, autonomous ChatGPT adjudication, concealed group-only forums, permanent automated sanctions, institution verification, private experimental workspace, or scientific claim promotion has been enabled.
Authenticated pending writes, one-level subtopics, private temporary holds, heading suggestions, moderator moves, reports, withdrawal, export, and production privacy screening are implemented.
Discussion remains separate from reviewed evidence, Recovery Points, and canonical model changes.