Content Moderation Policy

Effective Date: April 28, 2026

1. Moderation Points

ClipMagic reviews user prompts before generation, uploaded reference media before use, and generated outputs before delivery where technically available.

2. Automated Review

We use local keyword policies, OpenAI moderation for prompt review, and visual moderation providers for uploaded images and generated media. Visual checks may detect sexual content, minor safety risks, gore, weapons, offensive content, and public figure matches.

3. Video Review

Generated videos may be checked before delivery. Where supported, video review uses synchronous moderation and frame-level analysis. If a provider is unavailable in production, strict mode may block delivery instead of bypassing moderation.

4. Likeness and Voice Review

We may block or review prompts, uploads, and outputs that appear to recreate real people, public figures, celebrities, politicians, athletes, actors, creators, journalists, or recognizable voices.

5. Audit Logs

We keep structured moderation logs that may include user ID, timestamps, IP address, request IDs, prompt hashes, media URLs, provider decisions, categories, confidence scores, policy matches, and archived audit evidence where operationally required.

6. Outcomes

A request may be allowed, blocked, queued for review, or failed if a moderation provider is unavailable. Blocked content is not returned to the user, and credits for blocked or failed requests are released or returned where technically available.

7. User Violation Handling

First violations may result in content removal and a warning. Repeated violations may result in suspension, permanent account ban, IP restrictions, or payment instrument blocking. Severe violations may be reported to relevant authorities.

8. Transparency Review

We review moderation logs periodically, test edge-case prompts, monitor refund and dispute trends, and update this policy when enforcement controls materially change.

9. Appeals

Users may contact support@clipmagic.ai if they believe a moderation decision was incorrect. We aim to review moderation appeals within 5 business days.

Last reviewed: April 28, 2026.