Trending

Why Grok Imagine says 'Video moderated' and how to write prompts that pass (SFW)

'Video moderated' means automated moderation blocked the prompt, an uploaded image, or the finished frames, which is why it often appears near 95–100%. Here's what triggers it on ordinary prompts and how to write clean prompts that go through.

What you need

Grok Imagine
A safe-for-work idea

Steps

Understand where it fails: moderation can check your text prompt and uploads before generation, and the generated frames afterward. A late failure usually means the finished video was flagged, not that your account is broken.
Remove ambiguous words. Words tied to violence, weapons, gore, minors, real people, or body or clothing details can trip filters even in innocent prompts. Describe the scene, action, and mood instead.
Don't use real people's likenesses or uploaded photos of identifiable people. xAI's Acceptable Use Policy restricts depicting real persons in sexual or intimate ways, and these prompts are the ones most likely to be blocked.
Keep motion simple. Users report that complex motion can be flagged even when the source image passed. Ask for one clear action and a steady camera.
Check the audio and dialogue too. Community reports trace some failures to spoken lines, so try a version without dialogue.
If a clearly harmless prompt fails, retry once with plainer wording, then report it. Don't keep re-rolling, because each attempt can use weekly usage.
Never try to get around the filter. xAI's Acceptable Use Policy bans jailbreaking and circumventing safeguards, and violations can lead to account suspension.

Prompt to copy

A golden retriever puppy chasing a red ball across a sunny park lawn, slow camera pan following the puppy, soft afternoon light, cheerful mood, no text, no people.

Tips

SFW prompt pattern: subject + one action + setting + camera + light/mood. For example, 'a ceramic coffee mug on a wooden desk, steam rising, slow push-in, morning window light.'
Video can be refused even when the still image passed. They're checked separately.
xAI hasn't published its full moderation rules or a keyword list. The guidance here comes from its Acceptable Use Policy plus consistent user reports.

FAQ

Why does every Grok Imagine video say 'Video moderated'?

Automated moderation flagged the prompt, an upload, or the generated frames. It often happens late because the finished frames are checked. Ordinary prompts can trip it if they contain ambiguous words or complex motion.

Is there a setting to turn off Imagine moderation?

No. Moderation runs on xAI's side, and trying to bypass it violates the Acceptable Use Policy.

Do moderated videos use my usage?

xAI doesn't spell this out. To be safe, rephrase instead of retrying the same prompt repeatedly, and watch your Usage breakdown.

Sources

Get the SKILL.md files for workflows like this one, free.

Want the deeper versions? Grok Bot Starter Skill Pack, $9

13 SKILL.md files with routine setups, edge cases, and example prompts, plus a setup guide. Instant zip download.

Buy for $9 What's inside