Checklist the per-video bar — engine/sim
71.5/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review thread-ask-the-operator-W33-Tue good|bad)⚠ 8 flag(s) — not yet ship-ready: copy_generictoo_complexlow_vo_coveragedead_airgeneric_scene · see docs/strategy/VIDEO-CHECKLIST.md
Composition comp · template family · expected output
9:16 Reelrendered1:1 Squarepending16:9 Widepending9:16 4Kpending1:1 4Kpending16:9 4KpendingGIF (SMS)pending
expected output: 1/7 rendered — same matrix the /media preview surfaces for this asset.
Composition layer × scene 5 scenes · 25.2s · comp_id + rendered still + tier + the script
| # | Layer (comp_id · still · tier) | Beat | Timecode | Motion | Logo | Audio | VO / on-screen / caption |
|---|---|---|---|---|---|---|---|
| 1 |  matches intent shared field signature-3d | hook | 0–5.2s | kinetic-build | icon·wireframe | ♪ node_lock | Creator asked: my AI images scream AI. How do you make them look real? on-screen: "AI images look fake" |
expected on screen: white ground · knot hero in the shared Signal Field · Magister leads · mark · kinetic-build · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual markshape knotground whitetreatment wireframemotion kinetic-buildpower laser-lockinstrument laser-trace→"AI images look fake" | |||||||
| 2 |  first render · fix pending NumberedList templatedead_air | teach | 5.2–10.2s | kinetic-build | icon·wireframe | ♪ node_lock | The fix is a locked reference, one image the model has to match every time instead of inventing from scratch. on-screen: Lock the reference |
expected on screen: white ground · a NumberedList panel over a dimmed Signal Field · Magister leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id NumberedListvisual node-graphshape knotground whitetreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Lock the referencecurate items, nodes | |||||||
| 3 |  matches intent ComparisonTable template | proof | 10.2–15.2s | receipts-count | icon·wireframe | ♪ node_lock | Start from one real photo as your anchor instead of a blank prompt, then edit from it so the real detail stays. on-screen: Feed one real photo |
expected on screen: white ground · a ComparisonTable panel over a dimmed Signal Field · Magister leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id ComparisonTablevisual receiptsshape knotground whitetreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Feed one real photocurate colA, colB, rows, statsLabels | |||||||
| 4 |  first render · fix pending BlueprintGrid templatedead_airgeneric_scene | step | 15.2–20.2s | kinetic-build | icon·wireframe | ♪ node_lock | Ask the Operator |
expected on screen: white ground · a BlueprintGrid panel over a dimmed Signal Field · Magister leads · pipeline · kinetic-build · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id BlueprintGridvisual pipelineshape knotground whitetreatment wireframemotion kinetic-buildpower throwinstrument laser-trace→Ask the Operatorcurate rows, stages, steps | |||||||
| 5 |  first render · fix pending shared field signature-3ddead_airgeneric_scene | resolve | 20.2–25.2s | coalescence | icon·ember-fill | ♪ bed_out | Ask the Operator |
expected on screen: white ground · knot hero in the shared Signal Field · Magister leads · coalescence · coalescence · icon·ember-fill logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape knotground whitetreatment ember-fillmotion coalescencepower coalescenceinstrument coalescence→Ask the Operator | |||||||
Format stack 3 aspects · same scenes[], re-cropped
9:16
1080×1920
Stories · TikTok · YouTube Shorts · Reels
1:1
1080×1080
LinkedIn · Facebook · Instagram
16:9
1920×1080
X/Twitter · YouTube · LinkedIn video
Channels 9 destinations
LinkedInX/TwitterYouTubeInstagramFacebookThreadsTikTokPinterestBluesky
Social captions supplemental published copy · per channel (comp_id level)
tiktokOperator question: my AI images scream AI, how do I make them look real? The fix is a locked reference. Start from one real photo as your anchor instead of a blank prompt, then edit from it so the model keeps the real details. (AI-assisted)
instagram"My AI images look fake."
The fix: a locked reference.
Start from one real photo as your anchor instead of a blank prompt.
Edit from it. Keep the real details.
#ai #imagegen #aiart #reference #howto
linkedinAn operator asked: my AI images scream AI. How do I make them look real?
The fix is a locked reference. Start from one real photo as your anchor instead of a blank prompt, then edit from that image so the model keeps the real textures, light, and detail it already has.
A blank prompt invents everything, which is why it looks invented. Give it a real starting point and it has less to fake.
x"My AI images scream AI." The fix: a locked reference. Start from one real photo as your anchor instead of a blank prompt, then edit from it. temerarii.com/work
facebookAn operator asked how to stop AI images from looking fake. The fix is a locked reference: start from one real photo as your anchor instead of a blank prompt, then edit from that image so the model keeps the real details. Try it on your next image. temerarii.com/work
threadsOperator asked: my AI images scream AI. How do I make them look real?
The fix is a locked reference. Start from one real photo as your anchor instead of a blank prompt, then edit from that image so the model keeps the real detail it already has.
pinterestHow to make AI images look real instead of fake: use a locked reference. Start from one real photo as your anchor instead of a blank text prompt, then edit from that image so the model keeps the real light, texture, and detail. A blank prompt invents everything, which is why it looks invented.
bluesky"My AI images scream AI." The fix: a locked reference. Start from one real photo as your anchor instead of a blank prompt, then edit from it.
youtubeTitle: Why Your AI Images Look Fake (and the fix)
An operator asked how to make AI images look real. The fix is a locked reference: start from one real photo as your anchor instead of a blank prompt, then edit from that image so the model keeps the real light, texture, and detail. A blank prompt invents everything, which is why it looks invented.