0:00 / 0:00

Gunhild Johanne Reumert

@shedoesai · Digital creator

INSTAGRAM · 2025-08-04Source

2.0Klikes

4.0Kcomments

Remix This

Recreate with Runway Gen 4.5

Make your own AI viral video

Prompt

GLOBAL LOCK: 
Subject is a Caucasian female in her mid-30s with long blonde hair, wearing a simple white t-shirt and a black lapel microphone. She is positioned in a home studio with a dark bookshelf background, featuring warm practical lights and a prominent blue neon circular light. The video layout is a vertical 9:16 split-screen: the top 40% shows a dark-themed digital UI with a prompt input box, and the bottom 60% shows the subject sitting at a laptop. The lighting on the subject is a soft warm key light from the front-left. The color grade is vibrant with high contrast. Speech is clear, enthusiastic UGC-style narration.

[00:00–00:03]
Subject: Looking down at her laptop, lips moving as she narrates the prompt.
Environment: Top UI shows the text "Airpods dissassemble then becomes the apple logo" being typed into a dark search bar.
Action: Typing motion.
Camera: Static MCU of the subject; digital screen capture for the top.
Lighting: Warm studio light on subject; glowing UI elements.
Speech: Speaker A (On-camera): "Airpods disassemble then becomes the apple logo." Tone is curious and instructional.

[00:03–00:07]
Subject: Looks up at the top screen, eyes wide, mouth open in a "Wow" expression. She brings both hands up to the sides of her head in a gesture of amazement.
Environment: Top video shows white AirPods exploding into internal circuitry and components, which then swirl together in a trail of light to form a glowing white Apple logo against a black background.
Action: High-energy reaction.
Camera: Static MCU.
Lighting: The glow from the Apple logo reflects slightly on the subject's face.
Speech: Speaker A (On-camera): "Wow!" High pitch, expressive.

[00:07–00:11]
Subject: Looks back down at the laptop, typing again.
Environment: Top UI shows the text "A black tesla box opens then becomes a tesla showroom" being typed.
Action: Focused typing.
Camera: Static MCU.
Speech: Speaker A (On-camera): "A black tesla box opens then becomes a tesla showroom." Steady, rhythmic cadence.

[00:11–00:16]
Subject: Subject smiles and nods, looking at the top screen. She raises her hands again, gesturing toward the screen.
Environment: Top video shows a sleek black box with a glowing white Tesla 'T' logo. The box unfolds mechanically to reveal a brightly lit, minimalist white Tesla showroom with a car in the center and charging stations.
Action: Satisfied reaction.
Camera: Static MCU.
Lighting: Bright white light from the generated showroom video.
Speech: Speaker A (On-camera): "Crazy!" Short, punchy emphasis.

[00:16–00:21]
Subject: Typing the final prompt.
Environment: Top UI shows the text "A Jeep of box opens up in the desert" being typed.
Action: Quick typing.
Camera: Static MCU.
Speech: Speaker A (On-camera): "A Jeep of box opens up in the desert."

[00:21–00:26]
Subject: Subject transitions to a full-screen view (or stays in split-screen with the Jeep video playing). She looks directly into the lens, smiling broadly and using expansive hand gestures.
Environment: Top video shows a rugged, military-style crate in a sandy desert at sunset. The crate opens to reveal a dark green Jeep SUV.
Action: Direct-to-camera address, enthusiastic hand waving.
Camera: MCU, subject centered.
Lighting: Consistent warm/cool studio contrast.
Speech: Speaker A (On-camera): "Insane! Comment WOW and I'll send over a free guide on how you can create these videos." Fast pace, high energy, clear CTA.

NEGATIVE PROMPT:
Visual: blurry faces, inconsistent hair color, flickering background lights, distorted logos, messy UI text, jittery camera movement, unnatural hand anatomy, low resolution, dull colors.
Speech: robotic voice, monotone delivery, background noise, muffled audio, lip-sync mismatch, stuttering, flat emotions.

SPEECH PACK:
[00:00-00:03] "Airpods disassemble then becomes the apple logo."
TAKE_A: (Curious) "Airpods... disassemble... then becomes the Apple logo."
TAKE_B: (Fast) "Airpods disassemble then becomes the Apple logo!"
TAKE_C: (Instructional) "Airpods disassemble, then becomes the Apple logo."

[00:03-00:07] "Wow!"
TAKE_A: (Gasp) "Wow!"
TAKE_B: (Long) "Woooow!"
TAKE_C: (Whisper) "Wow."

[00:07-00:11] "A black tesla box opens then becomes a tesla showroom."
TAKE_A: (Excited) "A black Tesla box opens, then becomes a Tesla showroom!"

[00:21-00:26] "Insane! Comment WOW and I'll send over a free guide on how you can create these videos."
TAKE_A: (High Energy) "Insane! Comment WOW and I'll send over a free guide on how YOU can create these videos!"
Prosody: Emphasis on "WOW" and "YOU". Short pause after "Insane!".

How shedoesai Made This Split Screen Transformation Tutorial AI Video — and How to Recreate It

This case study analyzes a high-performing "AI Reaction" video by creator @shedoesai. The video leverages a split-screen layout to demonstrate the power of generative AI video tools. The top half features a sleek, dark-mode UI where text prompts are typed in real-time, followed by cinematic product transformations (AirPods to Apple logo, Tesla box to showroom, Jeep box to desert vehicle). The bottom half features the creator reacting with high-energy expressions and gestures. This cinematic editorial portrait style, combined with warm studio lighting and a UGC aesthetic, creates a perfect balance between high-end tech demonstration and relatable human connection. With over 4,000 comments, the video's primary success lies in its "comment-to-get-guide" automation strategy.

What You’re Seeing: A Visual Breakdown

The video is structured as a series of three "reveal" beats. Each beat follows a consistent rhythm: typing the prompt, showing the AI generation, and the creator's reaction. The lighting is professional yet cozy, featuring a warm key light on the creator and a distinct blue neon practical light on the bookshelf behind her, adding depth and a "tech-savvy" vibe to the scene.

Shot-by-Shot Breakdown

Time Range	Visual Content	Shot Language	Lighting & Color	Viewer Intent
00:00–00:07	AirPods disassembling into the Apple logo. Creator reacts with "Wow."	Split-screen; Static MCU (Creator) / Digital UI (Top)	Warm skin tones; High-contrast black/white AI video.	Hook: Immediate visual "magic" to stop the scroll.
00:07–00:16	Tesla box unfolding into a futuristic showroom. Creator says "Crazy."	Split-screen; Same framing.	Cool blue background accents; Bright white showroom interior.	Reinforce: Proving the AI's consistency and quality.
00:16–00:21	Jeep box opening in a dusty desert. Creator says "Insane."	Split-screen; Same framing.	Golden hour desert tones; Warm studio lighting.	Escalation: Showing variety in environments (indoor vs. outdoor).
00:21–00:26	Full-screen creator talking directly to camera.	Medium Close-Up (MCU)	Consistent warm/cool contrast.	Conversion: Clear CTA to drive engagement and leads.

Why It Went Viral: The "Magic Box" Effect

The Power of Visual Satisfaction

This video taps into the "Oddly Satisfying" psychological trigger. Watching complex objects like AirPods or a Tesla box disassemble and reassemble into something else provides a high level of visual dopamine. By using recognizable global brands (Apple, Tesla, Jeep), the creator instantly anchors the viewer's interest in familiar, high-status imagery. The "Magic Box" concept—where a simple container unfolds into a complex world—is a classic storytelling trope that works exceptionally well for short-form video because it creates a clear "before and after" within seconds.

Platform Signals & Engagement Loops

From a platform perspective, the video is engineered for watch time and comments. The fast-paced typing (0–3 second hook) creates a "wait for it" effect. The creator’s reactions (hands to head, wide eyes) serve as social proof, telling the viewer that what they are seeing is indeed impressive. Most importantly, the "Comment WOW" call-to-action is a masterclass in engagement hacking. It reduces the friction for the user to comment, and the high comment count signals to the Instagram algorithm that the content is highly engaging, pushing it to a wider audience.

5 Testable Viral Hypotheses

The Split-Screen Authority: Showing the "input" (prompt) and "output" (video) simultaneously builds immediate trust and tutorial value.
Brand Association: Using logos like Apple and Tesla captures the attention of fans of those brands, increasing shareability.
The Reaction Mirror: Human reactions in the frame increase the viewer's own emotional response to the visual.
Low-Friction CTA: Asking for a specific, short word ("WOW") results in 10x more comments than a generic "What do you think?"
Visual Contrast: The dark UI of the prompt box contrasts sharply with the bright, colorful AI results, keeping the eyes engaged.

How to Recreate: From 0 to 1

Topic Selection: Choose 3 high-status brands or objects. Think of a "transformation" for each (e.g., a coffee bean turning into a Starbucks store).
Character Consistency: Set up your "reaction" space. Use a consistent background (like a bookshelf) and lighting (warm key light + cool accent light).
Generate AI Clips: Use a tool like Luma Dream Machine, Kling, or Runway Gen-3. Use prompts that specify "unfolding," "disassembling," or "transforming."
UI Overlay: Create a simple dark-mode text box graphic. You can use Canva or CapCut to animate the text "typing" in.
Recording the Reaction: Record yourself sitting at your laptop. Watch the AI clips on your screen and react naturally. Don't overact, but be expressive.
The Edit: Use a 9:16 canvas. Place the AI UI/Video in the top 40% and your reaction in the bottom 60%. Sync the "reveal" of the AI video with your verbal reaction.
Sound Design: Add a "typing" sound effect for the prompts and a "whoosh" or "cinematic hit" when the video transforms. Use a trending, upbeat tech-style BGM.
Automation Setup: Use a tool like ManyChat. Set a trigger so that when someone comments "WOW," they automatically receive a DM with your "guide" or link.

Growth Playbook: Distribution & Scaling

3 Ready-to-Use Opening Hooks

"I can't believe AI generated this in 30 seconds..."
"Stop scrolling! You need to see this AI transformation."
"The secret to cinematic AI videos is finally out."

4 Caption Templates

The Value Lead: "AI is getting scary good. 🤯 I just turned a simple box into a Tesla showroom using one prompt. Want the workflow? Comment 'WOW' and I'll DM it to you! 👇"
The Curiosity Gap: "Which transformation was your favorite? 1, 2, or 3? 🎥 I'm sharing my exact prompt list for free today. Just comment 'WOW' to grab it!"
The Tutorial Tease: "You don't need a film crew anymore. Just an idea and the right AI tools. 🛠️ I've put together a full guide for beginners. Comment 'WOW' for access."
The Short & Punchy: "The future of content creation is here. 🚀 Comment 'WOW' for the full free guide on how to make these!"

Hashtag Strategy

Broad (Reach): #ai #artificialintelligence #contentcreation #digitalmarketing
Mid-Tier (Niche): #aivideo #runwayml #lumaai #videotips
Long-Tail (Specific): #indiecreator #aiartcommunity #aitutorial #socialmediagrowth

Frequently Asked Questions

What tools make it look the most similar?

Runway Gen-3 or Luma Dream Machine are best for these high-end product transformations.

What are the 3 most important words in the prompt?

"Cinematic," "Transform," and "Hyper-realistic" are key for this specific aesthetic.

How can I avoid making it look like AI?

Use high-quality reference images (Image-to-Video) to maintain brand logo accuracy.

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

Instagram currently favors high-aesthetic "tech-lifestyle" content like this more than TikTok.

How should I properly disclose AI use?

Use the platform's built-in "AI-generated" label and mention it in your caption to build transparency.