0:00 / 0:00

▶

Comment “UGC” for the AI workflow👇 Want to create AI UGC content at scale with realistic AI characters? With a workflow like this you can create infinite videos with infinite products. #weavyai #aiworkflow #aiautomation #aiugc

Gunhild Johanne Reumert

@shedoesai · Digital creator

INSTAGRAM · 2025-11-10Source

1.1Klikes

3.3Kcomments

Remix This

Recreate with Runway Gen 4.5

Make your own AI viral video

Prompt

GLOBAL LOCK:
Subject: Caucasian female, mid-20s, long blonde hair with soft waves, blue eyes, fair skin with warm undertones. 
Wardrobe: Simple beige or light pink silk camisole. 
Environment: Indoor bedroom setting, soft out-of-focus background with plants and warm interior elements. 
Lighting: Strong, warm directional sunlight from the side (golden hour), creating soft shadows and high-end cinematic highlights on the face. 
Color Grade: Warm, saturated tones, high contrast, clean highlight rolloff. 
Camera: Static Medium Close-Up (MCU), 50mm lens feel, shallow depth of field. 
Speech Signature: Friendly, conversational female voice, medium pace, warm and trustworthy tone.

[00:00–00:04]
Subject: The blonde woman is holding a small pink jar (Laneige Lip Sleeping Mask) near her right cheek.
Action: She is looking directly into the lens, speaking with a natural smile. Her head tilts slightly to the left as she begins her sentence.
Speech: "This is my secret to waking up with smooth, hydrated lips."
Lip-Sync: High strictness; mouth movements must perfectly match the words "secret," "smooth," and "hydrated."
Motion: Subtle hair movement as if from a light breeze, natural blinking, and micro-expressions around the eyes.

[00:04–00:08]
Subject: Same woman, same position holding the pink jar.
Action: She continues speaking, maintaining eye contact. A slight nod of the head for emphasis on the word "works."
Speech: "This mask works while I dream."
Lip-Sync: High strictness; mouth must close on the "m" in "dream."
Motion: The jar remains steady in her hand. The warm light creates a slight shimmer on her lips.
Transitions: Continuous shot, no cuts. The video ends on a friendly, closed-mouth smile.

NEGATIVE PROMPT:
Visual: Robotic movement, distorted fingers, floating product, flickering light, blurry face, inconsistent hair texture, uncanny valley eyes, double chin, warped background.
Speech: Robotic cadence, monotone delivery, muffled audio, lip-sync delay, harsh 's' sounds, unnatural pauses, background noise, clipping.

SPEECH PACK:
Transcript:
00:00-00:04: "This is my secret to waking up with smooth, hydrated lips."
00:04-00:08: "This mask works while I dream."

TAKE_A (Natural/Influencer): [Warm, slightly upbeat] "This is my secret... to waking up with smooth, hydrated lips. This mask works... while I dream."
TAKE_B (Soft/ASMR-lite): [Whispery, intimate] "This is my secret to waking up with smooth, hydrated lips. This mask works while I dream."
TAKE_C (Direct/Authoritative): [Clear, punchy] "This is my secret to waking up with smooth, hydrated lips. This mask works while I dream."

Prosody Markup:
"This is my **secret**... to waking up with **smooth**, hydrated lips. This mask **works**... while I **dream**."
(Pauses at "..." for natural breath)

How shedoesai Made This AI UGC Workflow AI Video and How to Recreate It

This case study analyzes a high-performing AI-generated UGC (User Generated Content) video for a skincare product. The video features a hyper-realistic blonde female character in a cinematic bedroom setting, illuminated by warm, directional "golden hour" light. By combining a static AI character image with a real product asset and advanced lip-syncing technology, the creator demonstrates how to bypass traditional filming costs while maintaining the "authentic" feel of a beauty influencer review. The core keywords for this aesthetic are cinematic editorial portrait, AI UGC, warm bedroom lighting, and product integration.

What You’re Seeing: A Visual Breakdown

The video utilizes a split-screen comparison layout to build trust and curiosity. On the left, we see the "ingredients": a high-quality AI-generated portrait and a clean product shot of the Laneige Lip Sleeping Mask. On the right, the final composite video shows the character holding the product, speaking naturally with perfect lip-syncing. The wardrobe is a simple, neutral-toned camisole that doesn't distract from the face. The lighting is the standout feature—a strong, warm key light from the side creates a "sun-kissed" look with soft shadows, mimicking a high-end iPhone vlog. The music is a low-fi, upbeat track that complements the "morning routine" vibe.

Shot-by-Shot Breakdown

Time Range	Visual Content	Shot Language	Lighting & Tone	Viewer Intent
00:00–00:03	Character introduces the "secret" while holding the jar.	Medium Close-Up (MCU)	Warm, golden hour, high contrast.	Hook: Establish problem/solution.
00:03–00:06	Character explains the benefits (smooth, hydrated lips).	MCU, slight head tilt.	Consistent warm glow.	Reinforce persona & product value.
00:06–00:08	Closing statement about the mask working while dreaming.	MCU, friendly smile.	Soft highlight rolloff on skin.	CTA/Closing: Emotional payoff.

Why It Went Viral: The Mechanics of AI Realism

The Power of "UGC" Aesthetics

This video taps into the "UGC" (User Generated Content) trend, which currently outperforms traditional high-production ads on platforms like Instagram and TikTok. By making the AI character look like a real person filming in their bedroom, it bypasses the "ad blindness" most users have. The choice of a popular, recognizable product (Laneige) provides an immediate anchor of reality for the viewer.

Psychology of the "Secret" Hook

The script starts with "This is my secret to..." which triggers a biological curiosity response. People are naturally inclined to stop scrolling when they feel they are about to learn a hidden tip or a shortcut to a desired outcome (in this case, perfect lips). The high visual quality removes the "uncanny valley" friction that usually kills AI video engagement.

Platform Signal Analysis

From a platform perspective, the video likely triggered high save rates. Creators and marketers save this content as a reference for their own AI workflows. The "Comment 'UGC' for the workflow" call-to-action (CTA) in the caption is a classic engagement hack that signals to the algorithm that the content is highly interactive, boosting its reach significantly.

5 Testable Viral Hypotheses

Hypothesis 1: The "Real Product" Anchor. By showing a real, physical product next to the AI, you ground the digital creation in reality, increasing trust. Replicate by: Overlaying a PNG of a real product in your AI video.
Hypothesis 2: Golden Hour Lighting. Warm, directional light is universally perceived as "aesthetic" and "high-end." Replicate by: Adding "golden hour, warm side lighting" to your image prompts.
Hypothesis 3: The Split-Screen "How-To." Showing the process (AI image + product = video) provides immediate educational value. Replicate by: Using a 3-part grid layout for your hook.
Hypothesis 4: Perfect Lip-Sync. High-fidelity lip-syncing removes the "AI-ness" and keeps viewers focused on the message. Replicate by: Using tools like Hedra or Sync Labs for 1:1 mouth movement.
Hypothesis 5: Niche Authority. Focusing on a specific niche (Skincare/Beauty) attracts a high-intent audience. Replicate by: Choosing one specific industry and sticking to its visual tropes.

How to Recreate: From 0 to 1 Workflow

Step 1: Character Positioning

Identify your niche. For beauty, you need a character with clear skin and relatable features. Use Midjourney to generate a consistent character portrait using a --cref (Character Reference) tag.

Step 2: Environment & Lighting

Ensure your prompt includes specific lighting instructions. Use: "Shot on iPhone, bedroom background, warm sunlight through window, cinematic depth of field."

Step 3: Product Integration

Take a high-quality photo of your product or find a clean PNG. Use Photoshop or Canva to place the product in the character's hand in your base image. This "composite" will be the foundation for your video.

Step 4: Video Generation

Upload your composite image to an AI video generator like Kling AI or Luma Dream Machine. Use a prompt like: "Woman holding a jar, talking to camera, subtle head movements, blinking, realistic skin texture."

Step 5: Audio & Scripting

Write a 10-15 second script. Record your own voice or use an AI voice generator (ElevenLabs) with a "natural, conversational" tone. Avoid sounding like a radio announcer.

Step 6: Lip-Syncing

Take your generated video and your audio file into a tool like Hedra or LivePortrait. This will sync the character's mouth movements perfectly to the speech.

Step 7: Final Edit & Overlays

In CapCut, add the "AI Image" and "Real Product" overlays as seen in the original video. Add bold, sans-serif captions at the bottom to highlight the "How-to" aspect.

Step 8: Publishing Strategy

Post as an Instagram Reel or TikTok. Use a cover image that clearly shows the "Before/After" or the "Process" to encourage clicks.

Growth Playbook: Distribution & Scaling

3 Ready-to-Use Opening Hooks

"Stop filming UGC. Do this instead..."
"My secret to 10x-ing my content output with AI."
"How I created this viral skincare ad in 15 minutes."

4 Caption Templates

The "Value-First" Template:
[Hook: Want to create AI UGC at scale?] 👇
[Value: This workflow lets you swap products and characters in seconds.]
[Engagement: Which product should I try next?]
[CTA: Comment 'WORKFLOW' for the full guide!]

Hashtag Strategy

Broad (Reach): #ai #contentcreation #digitalmarketing
Mid-Tier (Niche): #aiugc #aiautomation #skincaremarketing
Long-Tail (Specific): #weavyai #aiworkflowtutorial #realisticai

Frequently Asked Questions

What tools make it look the most similar?

Midjourney for the base image, Kling AI for motion, and Hedra for lip-syncing.

What are the 3 most important words in the prompt?

"Cinematic," "Golden hour," and "Photorealistic."

Why does the generated face look inconsistent?

You need to use a Character Reference (cref) in Midjourney to lock the facial features.

How can I avoid making it look like AI?

Add "film grain" and "slight camera shake" in post-production to mimic handheld filming.

Is it easier to go viral on Instagram or TikTok with this?

Instagram currently favors high-aesthetic "educational" AI content like this.