
#1 Platform to Create AI UGC - https://xitter.cf/t.co/G1asa9294y
Try Now →
Joined August 2024
- Tweets2.1K
- Following10
- Followers9.7K
- Likes3.5K
Pinned Tweet
We Just Made Making 100+ Ads - Simple.
No Creators
No Cameras
No Endless Edits
@makeugc just destroyed a $7 Billion industry overnight with 1 AI agent
Creating ads is now like stealing candy from a baby
Upload your ad and recreate it 100% with AI in minutes
Start creating your ads now on MakeUGC
MakeUGC retweeted
Your AI UGC creator looks different in every video. Here’s the fix.
I shipped 300+ creatives with Claude + GPT-Image-2.5 + Seedance 2.5, and the biggest problem was never quality. It was consistency.
Same prompt. Same reference. A different person every time:
- The jawline shifts
- The hair falls differently
- The freckles move 2mm
Small stuff, but across scenes and ads, people notice.
The fix is a system, not a better prompt. Here’s what’s inside:
✓ Identity sheet
- Locks one face across every render
- Works this week and next month
✓ 6 copy-paste prompts
- Angle minerIdentity sheet
- Hook script
- Product-in-hand render
- Variation matrix
- Reject pass
✓ Casting table
- 6 creator archetypes
- Matched to 6 DTC verticals
✓ Master render prompt
- Bracket-filled for skincare, supplements, apparel, home, food amp; bev, and accessories
✓ 6-check reject pass
- Kills a bad take before it reaches your ad account
✓ Approval table
- One row, you fill it in, done
All you need is one product photo.
I packaged the whole process into one guide.
Get free access below. 👇
Paid ads are cooked 😭
Seedance 2.5 on MakeUGC it’s a killer.
And you cannot even tell it’s AI anymore...
I literally made 10 ads for under $5
→ Insane realism, from faces to hands to motion
→ Studio-level quality without a studio budget
The money making opportunities
with this are endless.
You can literally produce hyper-realistic UGC
winning ads for less than $1
with Seedance 2.5 on MakeUGC.
MakeUGC retweeted
I just cracked AI ads — and the numbers are ridiculous.
Built a pipeline that turns any flat product photo into a studio-quality video ad, using Claude as the director and MakeUGC as the engine.
What it delivers:
→ Hook that lands the problem in 2 seconds
→ Cost that's pennies vs. traditional UGC
→ Scale — 15x your creative output overnight
The flow: product image → video generation → AI voiceover → music → captions. One automated pipeline, start to finish.
Comment "AI" and I'll DM you the full breakdown
MakeUGC retweeted
Just made this with Seedance 2.5. Full workflow, start to finish:
- Grab a reference image — TikTok, Instagram, or Pinterest
- Upload it to Claude, write a JSON prompt, adjust as you like
- Generate the image
- Upload it to Seedance 2.5 with your full video prompt
Done
All inside MakeUGC, for $1.
I made a full guide breaking down the exact JSON prompt structure that makes this work.
Comment "JSON" and I'll send it over
🤖 Made with AI
I cracked the code.
Most people struggle with AI video prompts.
The trick isn't writing more — it's writing structured. I use Claude to turn a simple idea into a full JSON-style prompt: scene, character, camera movement, timing, negative prompts, all mapped out.
That structure is the difference between "obviously AI" and something that looks real.
Once you have the format, it's stupidly easy to make consistently realistic videos, over and over.
Want the exact JSON prompt structure I use?
Comment "WANT" and I'll send it over
MakeUGC retweeted
I cracked the code.
Most people struggle with AI video prompts.
The trick isn't writing more — it's writing structured. I use Claude to turn a simple idea into a full JSON-style prompt: scene, character, camera movement, timing, negative prompts, all mapped out.
That structure is the difference between "obviously AI" and something that looks real.
Once you have the format, it's stupidly easy to make consistently realistic videos, over and over.
Want the exact JSON prompt structure I use?
Comment "WANT" and I'll send it over
🤖 Made with AI
Seedance 2.5 just went unlimited on MakeUGC.
Unlimited videos, no per-generation cost
- 100/200 free credits
- Every model included
- 10 min processing per video Limited time only.
If you've got ad ideas sitting untested, nows the window.
Im also sending a prompt guide + template guide with it.
🤖 Made with AI
MakeUGC retweeted
Everyone said nobody rides a motorcycle from the handlebars. Watch what she does at 0:06.
Thanks to @makeugc for letting me recreate this from Seedance 2.5. I’ll leave a step-by-step guide in the comments.
Prompt + guide in the replies👇
🤝 Paid partnership · 🤖 Made with AI
MakeUGC retweeted
How soon will traditional product commercials with celebrities, professional cameras, and huge production crews become a thing of the past?
With unlimited Seedance 2.5 on @makeugc, you can create a product commercial in just 10 minutes without a production budget.
10 minutes for the script. 10 minutes for generation.
No massive production crews. No insane budgets. All you need is a couple of reference images and a prompt created with ChatGPT.
Want to try it yourself? You can get started for just $1!
Link in the comments.
PROMPT⏬:
Create a 20-second, horizontal 16:9, 1080p ultra-realistic cinematic luxury mascara commercial. Premium cinema-camera quality, sophisticated lighting, realistic materials, and precise, dynamic camera movements. The visual style combines deep blacks, dark reflective water, polished surfaces, and elegant beauty lighting.
## REFERENCE IMAGES
* **@Image-1:** Exact mascara product reference. Preserve the original packaging, proportions, colors, cap, surface finish, and visible branding.
* **@Image-2:** Camera composition reference for the woman's face during mascara application. Match the general viewing angle and framing.
## GENERAL VISUAL STYLE
High-end international beauty advertising. Dark, elegant, dramatic, and luxurious. Use realistic skin texture, controlled highlights, cinematic depth of field, natural motion blur, and sophisticated studio lighting.
Camera movements must be clearly visible and physically motivated. Use smooth orbital movement, low-angle tracking, a fast rising orbit, and precise close-up framing. Avoid static shots when movement is specified, digital zooms that imitate physical movement, and abrupt unintended camera changes.
## LOCATION
**ENVIRONMENT 1 — DARK WATER**
A nearly black water surface in a dark studio. The mascara lies horizontally on the water, surrounded by darkness. Soft directional lighting reveals the packaging and creates subtle reflections. Small concentric ripples spread naturally around the product.
**ENVIRONMENT 2 — MIRRORED BEAUTY STUDIO**
A dark, sophisticated room with tall mirrors and glossy black surfaces. A narrow runway extends through the center, level with the surrounding dark water.
A beautiful adult woman, approximately 26–28 years old, wears an elegant black evening dress. She has realistic skin texture, expressive eyes, and naturally beautiful eyelashes.
## MUSIC AND AUDIO
A sophisticated electronic soundtrack with a dark luxury atmosphere. Begin with subtle water ambience, then introduce a deep electronic pulse as the model appears. Build intensity during the camera maneuver, with a sharp musical accent on the eye close-up and a powerful, refined impact during the final product reveal.
Subtle water ripples, footsteps, and mascara application sounds. No dialogue or narration.
## SHOT SEQUENCE
### 00:00–00:03 — THE PRODUCT ON WATER
A close cinematic view of dark water reveals the mascara from @Image-1 lying horizontally on the surface. Its packaging is illuminated by a narrow spotlight, with realistic reflections beneath it.
Small concentric ripples spread outward around the stationary product.
The camera performs a smooth orbit through approximately one-third of a full circle around the mascara, physically changing its viewing angle and revealing different sides of the packaging. Keep the product recognizable and stable.
End with a sharp cinematic transition.
### 00:03–00:07 — THE MODEL ON THE RUNWAY
Reveal a dark mirrored room. A narrow black runway extends through the center, perfectly level with the surrounding dark water.
The camera is positioned extremely low, almost at floor level, looking along the runway.
The woman in the elegant black evening dress walks confidently toward the camera. Her movement is graceful and natural, and the dress responds realistically to her steps. Tall mirrors reflect the model, runway, and lights with coherent perspective.
The camera tracks backward at low height. The woman stops naturally on the runway.
### 00:07–00:12 — A DYNAMIC ORBIT TO HER FACE
As the woman stops, the camera performs a sudden, fast, controlled vertical maneuver.
**Starting near runway level, the camera rapidly rises toward the woman's head height while simultaneously moving in an arc behind her. It then completes the orbit around her shoulder and sharply decelerates, stopping in front of her face at a slightly off-center side angle.**
This must be a genuine physical camera movement combining a vertical rise and a curved orbit. Do not replace it with a simple tilt, digital zoom, or stationary rotation.
The camera briefly passes behind the woman's silhouette, creating a strong sense of speed and spatial movement. Use realistic motion blur during acceleration and a precise, stable stop.
The final framing matches the general angle shown in @Image-2. The woman's face and eyelashes are clearly visible.
She raises the mascara and applies it to her upper eyelashes in one controlled, natural movement. Her hand and the mascara wand move realistically. Soft beauty lighting emphasizes her eyes and individual lashes.
### 00:12–00:15 — THE EYE REVEAL
A sharp cinematic transition reveals an extreme close-up of the woman's eye opening.
Her eye fills most of the frame. The eyelashes are long, dark, thick, voluminous, and beautifully separated, with realistic individual strands. They should look luxurious and dramatic without appearing artificial or excessively thick.
Soft directional lighting creates delicate highlights in the eye and along the lashes. Preserve natural skin texture and facial anatomy.
The eye opens naturally and confidently. Hold the close-up briefly to showcase the result.
### 00:15–00:20 — THE FINAL PRODUCT
Transition to a clean, dark product shot.
The mascara from @Image-1 stands vertically at the center of a glossy black surface, with a clear reflection beneath it. Preserve its exact packaging, proportions, colors, and branding.
A narrow spotlight emphasizes the product's contours and surface finish while maintaining a deep black background.
Behind the product, the following text appears:
**NUMBER ONE IN THE WORLD**
Use a bold, condensed uppercase typeface with elegant, premium styling. The text is positioned behind the product, not overlaid across its front. The mascara occupies the foreground and partially overlaps the lettering, creating realistic spatial depth while keeping the claim readable.
The camera performs a slow, controlled push-in. The mascara remains stationary, vertically aligned, and sharply focused. Subtle highlights reveal the packaging, while the reflection remains realistic.
End with a powerful, clean luxury beauty composition: the mascara in the foreground, the uppercase claim behind it, and an almost completely black background.
The music finishes with a deep, confident cinematic impact.
## CONTINUITY AND PHYSICAL REALISM
* Preserve the exact product design from @Image-1 and use @Image-2 as the reference for the facial angle.
* Maintain consistent model identity, facial anatomy, hairstyle, dress, and makeup.
* The mascara must rest naturally on the water, with believable contact, ripples, and reflections.
* The opening camera orbit must cover approximately one-third of a full circle.
* The runway remains level with the surrounding water. Mirror reflections must follow the scene's geometry.
* The woman walks naturally, stops clearly, and applies mascara with realistic hand and wand movements.
* The fast camera maneuver must combine a rapid vertical rise with a curved path behind and around the woman, ending in a stable, slightly off-center close-up.
* The eye close-up must show realistic, individually defined eyelashes and natural skin texture.
* The final product stands vertically and remains stationary. **NUMBER ONE IN THE WORLD** appears behind it in condensed uppercase lettering.
* Maintain coherent perspective, lighting, reflections, and object scale.
* Avoid duplicated or deformed products, distorted eyes or hands, extra fingers, flickering reflections, artificial water, excessive motion blur, and random environmental changes.
* No additional text, watermarks, or products.
## FINAL FEEL
A premium, cinematic mascara commercial combining dramatic product photography, dark reflective environments, elegant fashion imagery, dynamic camera choreography, and an extreme beauty close-up.
The finished video should feel sophisticated, modern, visually striking, and professionally directed, with realistic materials, precise camera movements, and a memorable final product reveal.
#AIVideo
🤝 Paid partnership · 🤖 Made with AI
MakeUGC retweeted
A girl warmed a freezing dragon, and he sheltered her from the blizzard. 🐉
Created with Seedance 2.5 and @makeugc.
Even those who seem unbreakable sometimes need a little warmth.
Share this video with someone who once warmed your heart with their kindness. ❤️
🤝 Paid partnership · 🤖 Made with AI
The only cardio that actually works.
Part 4 of Mila's gym diary.
Made with Seedance 2.5 on @makeugc, no queues, about 10 minutes per generation. Details in the comments.
How to repeat such a video?
The trick: describe recurring characters exactly the same way in every episode.
The trainer is from Part 3, and I copied his description word for word, so he looks like the same guy.
Prompt below 👇
🤝 Paid partnership · 🤖 Made with AI
She explains, with her mouth full, that calories don't count on weekends.
Then her trainer walks past the window.
Made with Seedance 2.5 on @makeugc. Details in the comments.
POMPT 🔽
SCENE:
A cheap fast food restaurant on a Sunday afternoon, a plastic table next to a big window. The young woman from @ Image1 and @ Image2, same face and dark brown hair, sits at the table. Her skin is dry and natural, no sweat. She is dressed up for a night out: a fitted black corset top with a deep neckline, high-waisted jeans, gold hoop earrings, long dark hair down and styled, evening makeup. In front of her is a huge tray of fast food: two burgers, a big box of fries, chicken nuggets, a large chocolate milkshake and a large cup with "DIET" written on it. She eats messily, completely ignoring how dressed up she is.
0-3s: she dips a handful of fries into the milkshake, stuffs them into her mouth, cheeks full, and says confidently to the camera: "Calories don't count on weekends."
3-6s: the friend behind the camera whispers: "Mila... your trainer." She freezes mid-chew, eyes wide, and slowly turns to the window. Outside, a fit man in a gym tracksuit stands and stares at her.
6-10s: in one quick move she shoves the whole tray across the table toward the camera, grabs a small cup of water, sips it and smiles innocently, waving at the window.
CAMERA:
Authentic casual footage captured with an older consumer digital camera. Handheld DV 16mm camcorder aesthetic. An unseen person operates the camera from a plausible nearby position. Preserve the viewpoint, framing and shot sequence described in the scene. Camera reactions feel human and slightly imperfect. The recording device stays outside the frame.
LOOK:
Soft digital-tape look with subtle vintage DV character. Soft fine detail, faint digital noise, mild compression, slightly muted colours, restrained contrast and imperfect white balance. Preserve natural skin tones, pores, wrinkles, fabric texture and realistic surface reflections. No beauty filters, plastic skin, glossy commercial finish or heavy colour grading.
HANDHELD:
Natural hand shake, slightly crooked framing and small framing corrections. Brief motion blur during camera movement, occasional delayed autofocus after reframing and subtle vertical movement from the operator's body. Keep faces and important actions readable. Avoid perfectly stabilized or gimbal-like movement.
LIGHTING:
Available daylight entering through the room's actual windows. Areas near the windows are brighter, with natural light falloff deeper into the room and soft reflections from nearby walls. Preserve gentle facial shadows, realistic skin tones and readable interior detail. No studio fill lights or artificial beauty lighting.
AUDIO:
Restaurant background noise, her muffled voice with a full mouth, the friend whispering, the tray sliding, the friend trying not to laugh.
He broke the wall. Then he saw the symbol. 💀
Made with Seedance 2.5 on @makeugc
15 seconds. One fight. A secret buried in the past.
Most AI anime fights have the same problem.
The punches look powerful, but nothing changes. Characters trade blows, walls explode, dust fills the screen - and the scene means nothing.
This one has a different purpose.
▪️ Vex throws a punch so close to Mara’s face that a few centimeters separate her from the impact.
▪️ She dodges. His fist destroys the wall behind her.
▪️ Mara counters with precision. Vex responds with pure aggression.
▪️ The damaged wall breaks open, revealing a forgotten mural - two silhouettes and a symbol neither of them can ignore.
▪️ Mara reveals the same mark on her wrist. Vex stops fighting.
Then comes the line that changes the entire scene:
You were there…
Suddenly, this isn’t a fight between strangers.
It’s a reunion neither of them was ready for.
The best twist doesn’t explain the whole story. It gives you one detail that makes you rethink everything you just watched.
✍️ The full prompt is in the comments.
🤝 Paid partnership · 🤖 Made with AI
MakeUGC retweeted
⚜️Little-known AI video lifehack: stop writing "small baby dragon" and write the size in centimeters.
Seedance 2.5 on @makeugc
Write "small baby dragon" and the model often renders something far bigger than intended. Words like small and tiny carry no scale, so it falls back on how dragons usually look.
It's not some secret hack. Give the creature a number, then put it next to an object the model already knows:
▪ "the size of a large cat, about 45 cm long and 25 cm tall at the shoulders"
▪ the cast-iron pot beside it: "35 cm wide"
▪ in the continuity block: "ONE dragon only, always SMALL, never large"
In this prompt the dragon leaps onto a wooden table and breathes fire into that pot. The pot works as a ruler, and the number says how big the dragon is next to it.
The model was never wrong about what a dragon looks like. It just had no ruler.
🔖Tomorrow, I’ll tell you something else that will significantly improve your generations.
🤝 Paid partnership · 🤖 Made with AI
Stop letting Seedance 2.5 improvise the transformed creature. Upload the result as its own reference
Seedance 2.5 on @makeugc . Prompt and the workflow👇:
Here is my advice to you:
When a character shows up at two sizes, a single picture isn't enough. The huge dragon and the baby end up with different colors, different horns, different eyes.
The fix is to give every form its own reference and assign the roles in the very first line:
"Use image 1 for the young woman's face, hair and outfit. Use image 2 for the adult dragon. Use image 3 for the baby dragon."
Three images, three jobs, one sentence.
Then tie the forms together in the continuity block: "The baby dragon has exactly the same pearl-white and pink color, amber slit-pupil eyes and pink-tipped crest as the adult dragon."
The dragon that fills the room and the one that fits in two hands are still clearly the same animal.
A dragon that shrinks is still one dragon. The prompt only has to say so.
MakeUGC retweeted
I wish GTA 6 had more interesting characters than normie dude and his girlfriend. Something like angry granny on a rampage 👵🏻
So I made this concept with Seedance 2.5 via @makeugc. Maybe @RockstarGames will add this lady to their game? Please?!
#gta6 #ai
🤝 Paid partnership · 🤖 Made with AI
MakeUGC retweeted
I recreated the iconic Transformers scene with a new Megan Fox using AI
Made with MiniMax Recast on @makeugc
Which Megan Fox do you prefer: 2007 or 2026?
🤖 Made with AI
She took his request a little too literally 💅
I turned a viral image meme into this video with @makeugc. Then I built the entire edit just by telling ChatGPT what I wanted.
Here’s the workflow:
1. Reference image: Nano Banana on MakeUGC.
2. Video: Wan 3.0 on the same platform. An affordable model with results like this.
3. Editing: ChatGPT 6 Astra added the WhatsApp notification, typing animation, reply, photo send, and sound effects. All from instructions in chat.
That’s what makes this exciting: a funny image can now become a realistic scene or a whole series for your account.
New characters, different situations, your own punchlines. Memes, sketches, short stories. AI makes it much faster and more affordable to create them, test what people respond to, and build an audience.
#AIVideo
MakeUGC retweeted
Of course AI gaming videos are NOT real games, that's just a VIDEO lol. But they can be a great tool as references for videogame devs
This action-slasher concept video was made with unlimited seedance on @makeugc, just around 5 minutes for generation
#AI
🤝 Paid partnership · 🤖 Made with AI
MakeUGC retweeted
When AI takes everyone’s job except yours 😭
Made with MiniMax Recast on @makeugc
Link in the comments.
#AIVideo
🤖 Made with AI
I made a video of what it would take for 2027 to top this year 💀
Pick your 2027 scenario.
Made with Seedance 2.5 on @makeugc
I built the prompt with ChatGPT 6 Astra, planning each scene, camera angle, lighting change, and reaction before generating the video.
To adapt it for your own video:
1. Swap the character, setting, or disasters.
2. Use the shot timings and camera directions as your starting point.
2. Match the character’s reactions to the chaos around them, and save a good punchline for the final shot.
The full prompt is almost 15,000 characters long, so I’ve put it in the comments. Copy it, change the details, and make your own version 👇
#AIVideo
The Mother of Dragons after a long day of burning cities.
Made with MiniMax on @makeugc. Details in the comments.
How to repeat?
Character swap with MiniMax Recast: one source video + character references.
2 tricks that finally made it work:
1. Tell the model to replace the HANDS too. Otherwise she gets the original guy's hands with his wedding ring
2. Tell it to copy every action "without skipping or softening". Otherwise it skips the funny parts and makes her just smile.
The ending is a separate Extend generation.
Prompt below 👇
🤝 Paid partnership · 🤖 Made with AI
He came to save her. She was having tea with the beast.
Never try to rescue the Mother of Dragons 🐉
Made with Seedance 2.5 on @makeugc. Details in the comments.
How to repeat?
А separate reference sheet for each main character (all angles + facial expressions), and every shot has only 1-2 actions. More than that and the model starts mixing things up.
Prompt below 👇
MakeUGC retweeted
Made with Seedance 2.5 on @makeugc
PROMPT ⬇️
Create a 15-second, ultra-realistic live-action beach vlog. Exactly 6 shots connected by 5 quick, clean hard cuts. All movement happens at normal speed.
CAMERA:
DV 16mm tape camcorder handheld feel. POV of MILA holding the camera herself, occasionally propping it on her beach bag for hands-free beach shots. Hand shake, misaligned framing, delayed focus pulls, clumsy zooms, occasional face cut-off framing, imperfect shots. Camcorder never appears on screen.
Selfie shots are filmed at a physically plausible arm’s-length distance. Propped-camera shots use ordinary fixed positions nearby. All camera repositioning happens between cuts, off-screen.
Keep the imperfections small and believable: irregular wrist movement, a brief accidental zoom adjustment, a moment of soft focus after a cut. No floating camera, smooth cinematic orbit or impossible third-person camera movement.
LOOK:
Soft, slightly blurry tape quality, faint tape noise, bloomed highlights under bright beach daylight, flickering auto-exposure, muted contrast, realistic skin tones.
Natural sunlight, soft reflected light from the sand, believable shadows and bright reflections on the sea. Preserve natural skin texture, subtle facial asymmetry and realistic hair detail beneath the soft tape image.
The footage should look like a real casual beach recording made with a consumer camcorder. No CGI appearance, waxy skin, beauty-filter smoothing or polished fashion-commercial lighting.
STYLE:
Playful, self-deprecating beach-vlog tone. MILA had planned to work out but has happily chosen to relax instead. Casual delivery, small amused expressions and slightly guilty smiles.
Quick handheld cuts, with the energy staying light and funny throughout. Humor comes from her reactions and dialogue. Physical actions remain simple and natural.
CHARACTER:
MILA — a beautiful American Instagram-style brunette model and performer, 20. Long dark-brown hair in a high ponytail, glowing skin with a light sweat sheen, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach and lean arms and legs.
She wears a sunshine-yellow supportive bikini top with thin straps and a deep scoop neckline showing cleavage, paired with matching yellow thong bikini bottoms. Opaque swimsuit fabric with a secure, consistent fit. Bare shoulders, arms, midriff and legs. Barefoot. She speaks with an American accent.
Preserve her face, age, hairstyle and specified body proportions throughout. Maintain her very large, full bust, very full rounded buttocks, narrow waist and lean limbs consistently across all camera angles. Natural anatomy and gravity appropriate to each pose.
SETTING:
A quiet sandy beach in warm afternoon daylight. A simple cream-colored sun lounger faces the sea at a slight angle. A folded towel rests on the lounger, with a beach bag nearby.
The shoreline is several meters away. Gentle waves remain in the background. A light breeze moves a few loose strands of her hair.
Keep the lounger, towel, bag, shoreline and sunlight direction consistent throughout.
STORYBOARD — 15 SECONDS, 6 SHOTS:
1. 0.0–2.5 seconds — ARM’S-LENGTH SELFIE
MILA is already standing beside the sun lounger, holding the camera slightly above eye level.
Frame her face, shoulders and upper torso. Her yellow bikini top and neckline are naturally visible.
She looks directly into the lens with a mildly guilty smile.
MILA: “I had a whole workout planned.”
Small wrist shake and a slightly imperfect framing correction.
HARD CUT.
2. 2.5–4.5 seconds — PROPPED CAMERA, REAR THREE-QUARTER VIEW
The camera is already propped securely nearby, behind MILA and slightly to one side.
A medium-wide shot frames her from her head to below her knees. She is already facing toward the lounger and sea. Her back, narrow waist, yellow thong bikini bottoms and full rounded buttocks are clearly visible within the overall composition.
She takes one small, relaxed step toward the lounger, then pauses. Ordinary walking posture and natural weight transfer.
No dialogue. Gentle surf and beach ambience.
HARD CUT.
3. 4.5–7.5 seconds — PROPPED CAMERA, FRONT THREE-QUARTER MEDIUM SHOT
MILA is already comfortably reclining on the lounger. Her sitting-down movement happens entirely between shots.
Frame her from her head to her upper thighs. Her face, yellow bikini top, waist and relaxed posture remain visible, with the sea behind her.
She settles her shoulders slightly, glances toward the water and gives the lens a pleased little smile.
MILA: “Then I found this spot.”
The fixed camera stays mostly steady.
HARD CUT.
4. 7.5–9.0 seconds — BRIEF HANDHELD FRAMING MISTAKE
Still reclining, MILA briefly aims the camera too low.
The framing runs from just below her chin to her waist, showing her upper torso, yellow bikini top, neckline and midriff. Most of her face is accidentally outside the top of the frame.
Small handheld wobble, ordinary breathing and a brief delay before autofocus settles. A tiny clumsy zoom correction, then the framing stops moving.
No dialogue. Natural beach sound only.
HARD CUT.
5. 9.0–11.5 seconds — PROPPED CAMERA, CLOSE MEDIUM SIDE ANGLE
MILA remains in the same reclining position.
A gentle three-quarter side angle frames her face, shoulders and upper torso. She looks toward the sea, briefly closes her eyes in contentment, then opens them with a small amused smile.
A few loose strands of hair move in the breeze. Natural breathing and subtle facial movement.
No dialogue. Let the quiet reaction carry the joke.
HARD CUT.
6. 11.5–15.0 seconds — ARM’S-LENGTH SELFIE FINISH
Still reclining, MILA holds the camera slightly above her face. Frame her face, shoulders and upper chest. The camera itself remains completely outside the image.
She looks directly into the lens, raises one eyebrow slightly and delivers the line with playful sincerity.
MILA: “I’m calling it active recovery.”
Hold her satisfied grin for the final beat. Slight hand shake, natural blinking and relaxed breathing.
AUDIO:
Natural American-accented speech with accurate lip synchronization. Her voice sounds close and clear, recorded through a consumer camcorder microphone.
Gentle surf, a light breeze and faint distant beach ambience. Slight wind noise during handheld shots, without covering her dialogue.
No music, narration, extra dialogue or added comedy sound effects.
CONTINUITY:
Exactly one MILA, with the same face, high ponytail, earrings, figure and yellow bikini in every shot.
Preserve the specified bust size and rounded buttocks without shrinking or reshaping them between cuts. Keep the bikini straps, neckline, thong cut, fabric and color consistent.
Natural hands, limbs, posture and weight distribution. Simple movements with no body warping or duplicated anatomy.
All camera changes occur through hard cuts. Camera setup and the transition from standing to reclining happen off-screen.
No mirrors, visible recording devices, captions, subtitles, logos or watermarks.
#AIVideo
🤖 Made with AI
Made with Seedance 2.5 on @makeugc (native 15-second video, no editing)
PROMPT ⬇️
**CAMERA:**
DV 16mm tape camcorder handheld feel. POV of MILA holding the camera herself, occasionally propping it on the floor or a mat for hands-free core shots. Hand shake, misaligned framing, delayed focus pulls, clumsy zooms, occasional face cut-off framing, imperfect shots. Camcorder never appears on screen.
**LOOK:**
Soft, slightly blurry tape quality, faint tape noise, bloomed highlights under gym lighting, flickering auto-exposure, muted contrast, realistic skin tones.
**STYLE:**
Playful, self-deprecating gym-vlog tone — genuine strain mixed with humor, complaining about core work while still pushing through it. Quick handheld cuts, energy staying light and funny throughout rather than fully exhausted.
**Character**
MILA — a beautiful American Instagram-style brunette model and performer, 20. Long dark-brown hair in a high ponytail, glowing skin with a light sweat sheen, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach and lean arms and legs.
She wears a cobalt-blue supportive sports bra with thin straps and a deep scoop neckline showing cleavage, paired with matching ultra-short fitted athletic shorts. Bare shoulders, arms and midriff. White ankle socks, sneakers off. She speaks with an American accent.
**Setting**
A gym mat area in the evening — mirror wall nearby, water bottle on the floor, soft overhead lighting, other equipment visible in the background.
**Storyboard (15s, 6 shots)**
1. *(~2.5s, propped camera, medium shot)* She lies down on the mat, propping up on her elbows, already sighing dramatically. MILA: “Core day. Pray for me.”
2. *(~2.5s, propped camera, plank position)* She holds a plank, arms shaking slightly, talking through gritted teeth. MILA (strained): “Why am I shaking already?”
3. *(~2.5s, medium propped shot, crunches)* She moves into a set of crunches, breathing hard, complaining between reps. MILA: “This looked way easier online.”
4. *(~2s, macro insert, shallow DOF)* Close-up on her hands gripping the mat edges during a leg raise, abs visibly working. No dialogue — ambient gym sound only.
5. *(~2.5s, handheld, leg raises)* She finishes a slow leg raise set, flopping back onto the mat immediately after, laughing at herself. MILA (breathless): “Yep—my abs just quit.”
6. *(~3s, arm’s-length selfie finish)* Still lying on the mat, she lifts the camera above her face (camera no visible), grinning tiredly. MILA: “Same time tomorrow? No promises.”
#AIVideo















