0:00 / 0:00

Kling’s New AI Model “O1” just released and it’s going to change how you edit videos. Comment “AI” for a link Kling O1 lets you upload any video and edit it with simple natural-language prompts. You can change lighting, clothing, objects, backgrounds — anything — with a single sentence. This is the future of AI VFX, and it’s going to completely transform how creators, editors, and filmmakers work. #KlingO1 #AIVideoEditing #AIVFX #GenerativeAI #CreativeAI

Rourke Sefton-Minns

@rourke · creator

INSTAGRAM · 2025-12-01Source

5.9Klikes

707comments

Remix This

Recreate with Kling 3

Make your own AI viral video

Prompt

GLOBAL LOCK: Male subject, mid-30s, brown hair and beard, wearing a white "Vans" custom classic baseball cap. The video style is a mix of raw UGC (User Generated Content) and high-end cinematic AI renders. Pacing is fast with cuts every 2-3 seconds. The color grade for AI shots is cinematic with high contrast and saturated tones. Speech is direct-to-camera, energetic, and instructional.

[00:00–00:02]
Subject: Male in black t-shirt and white cap.
Environment: Top half is a green screen studio; bottom half is a cinematic dark stone hallway (Harry Potter style).
Action: Subject points a wooden wand toward the camera. In the AI version, the wand tip glows with a bright blue light and he wears a Gryffindor-style robe.
Camera: Medium shot, static.
Lighting: Flat studio light (top), moody torchlight with blue accents (bottom).
Speech: "This AI update has changed visual effects forever."

[00:03–00:05]
Subject: Male, shirtless (top), wearing a 1920s wool suit and flat cap (bottom).
Environment: Top is a desert with a white RV; bottom is a dimly lit vintage room with a lamp.
Action: A makeup artist's hand applies powder to his face.
Camera: Close-up on the face.
Lighting: Harsh sunlight (top), warm, soft interior lighting (bottom).
Speech: "Because you can now create Hollywood level visual effects..."

[00:06–00:08]
Subject: Male in an orange and blue martial arts gi (Goku style).
Environment: Top is green screen; bottom is a rocky canyon under a starry night sky with two moons.
Action: Subject performs a "Kamehameha" pose, a glowing white energy ball forming between his hands.
Camera: Medium wide shot.
Lighting: Vibrant blue glow from the energy ball reflecting on the subject's face.
Speech: "...with one simple source video and a basic prompt."

[00:09–00:12]
Subject: Male in a white t-shirt and white cap.
Environment: A modern bedroom with a wooden wardrobe and a bed.
Action: Subject sits at a desk with a microphone, gesturing with his hands. The camera cuts between a front view and a "behind the shoulder" view of the same scene.
Camera: Medium shot, alternating angles.
Lighting: Natural window light.
Speech: "You can even create multiple camera angles from any scene now."

[00:13–00:16]
Environment: A winding road through a forest.
Action: A white SUV drives down the road. The scene seamlessly transitions from a lush green summer forest to a heavy snow-covered winter forest, then to a futuristic neon city at night.
Camera: High-angle tracking shot behind the car.
Lighting: Bright daylight to cool blue night with neon reflections.
Speech: "Not only that, but you can change the weather and environment..."

[00:17–00:19]
Subject: Male in a black tuxedo (James Bond style).
Environment: Top is green screen; bottom is a luxury modern villa overlooking the ocean at sunset.
Action: Subject holds a handgun, aiming it slightly off-camera with a serious expression.
Camera: Close-up, side profile.
Lighting: Warm golden hour light.
Speech: "...all in 1080p and 10 seconds long. So here's how you can do it too."

[00:20–00:40]
Visual: UI walkthrough of the Kling AI website. A cursor navigates through "Kling O1", "Video Generation", and "Transformation" tabs. It shows the process of uploading a video of a man hanging upside down and a reference image of Spider-Man.
Speech: Instructional narration explaining the steps: "Go to Kling, select O1 model, go to video generation, upload input video and reference image..."

[00:41–00:45]
Visual: Rapid montage of AI transformations: Subject as Obi-Wan Kenobi with a lightsaber, Jon Snow in the snow, Indiana Jones with a whip, and AI images of the subject with Billie Eilish, Taylor Swift, and The Rock.
Camera: Various medium and close-up shots.
Speech: "Kling's new update also comes with a brand new AI image model..."

[00:46–00:50]
Subject: Male in a white t-shirt and white cap.
Environment: A tropical beach at sunset with palm trees.
Action: Subject holds a movie clapperboard that says "TYPE AI IN THE COMMENTS". He snaps the clapperboard shut.
Camera: Medium shot.
Lighting: Warm, orange sunset glow.
Speech: "If you want to try it out for yourself, type AI in the comments and I'll send you the link."

NEGATIVE PROMPT: Unstable facial features, flickering clothing textures, distorted hands or fingers, blurry background elements in AI renders, robotic or desynchronized lip movements, low-resolution artifacts, unnatural lighting jumps between cuts, watermarks on AI generated segments.

SPEECH PACK:
[00:00–00:08]
TAKE_A: "This AI update has changed visual effects forever. Because you can now create Hollywood level visual effects with one simple source video and a basic prompt." (Energetic, fast-paced)
TAKE_B: "VFX just got a massive upgrade. You're looking at Hollywood quality results from just a simple phone video and a prompt." (Confident, slightly slower)

[00:46–00:50]
TAKE_A: "If you want to try it out for yourself, type AI in the comments and I'll send you the link." (Friendly, direct)
TAKE_B: "Want the link? Just comment 'AI' below and I'll drop it in your DMs." (Casual, punchy)

Why rourke's Kling O1 Video Editing Tutorial Went Viral

This case study analyzes a high-performing tutorial by creator @rourke, showcasing the capabilities of the Kling O1 AI model. The video leverages a "magic transformation" hook, where raw green-screen footage is instantly converted into Hollywood-level cinematic scenes. By blending UGC (User Generated Content) aesthetics with high-fidelity AI renders—ranging from Harry Potter-themed hallways to Dragon Ball Z energy blasts—the video captures the "wow factor" essential for short-form platforms like Instagram and TikTok. With over 5,800 likes and a high comment-to-like ratio (driven by an automated "Comment AI" CTA), this video serves as a masterclass in AI-assisted storytelling and engagement hacking for indie creators.

What You’re Seeing

The video is a fast-paced montage that alternates between "behind-the-scenes" (BTS) reality and "AI-rendered" fantasy. The subject is a male creator with a distinct beard and a white "Vans" baseball cap, providing a consistent anchor across wildly different environments. The lighting shifts from flat, green-screen studio light to moody, cinematic palettes (warm ambers for Peaky Blinders, cool blues for sci-fi). The editing uses a stacked split-screen layout to show the input and output simultaneously, making the AI's power undeniable.

Shot-by-Shot Breakdown

Time Range	Visual Content	Shot Language	Lighting & Tone	Viewer Intent
00:00–00:02	Harry Potter wand transformation.	Medium Shot (MS)	Green screen to dark stone hallway.	Hook: Instant visual payoff.
00:03–00:05	Peaky Blinders makeup transition.	Close-up (CU)	Natural desert light to warm interior.	Reinforce versatility of the tool.
00:06–00:08	Goku energy blast (Kamehameha).	Medium Wide (MWS)	Vibrant orange and blue glows.	Niche appeal (Anime/Pop culture).
00:09–00:12	Creator in bedroom explaining "Multiple Angles".	Talking Head / POV	Natural window light, soft.	Tutorial value: Explaining features.
00:13–00:16	Car driving through changing weather.	Wide Shot (WS) / Tracking	Lush green to snowy white.	Showcase environment control.
00:20–00:40	UI Walkthrough of Kling O1 interface.	Screen Recording / Overlay	Digital interface.	Educational: "How-to" proof.
00:46–00:50	Creator on beach with clapperboard.	Medium Shot (MS)	Golden hour sunset.	CTA: Drive comments and engagement.

Why It Went Viral

The "Magic Mirror" Psychology

This video taps into the fundamental human desire for transformation. By showing the "ugly" green screen version alongside the "perfect" cinematic version, it creates a contrast loop. Viewers are naturally inclined to re-watch to see how specific details (like the beard or the hat) are preserved through the AI transformation. This increases "Watch Time," a primary signal for the Instagram algorithm.

The "Comment for Link" Engagement Hack

The creator uses a "Lead Magnet" strategy. By asking users to "Comment AI," he triggers two things: 1) A massive spike in the comment count, which signals to the platform that the content is highly engaging, and 2) An automated DM (likely via ManyChat) that delivers the link, moving the user further down the conversion funnel. This turns a passive viewer into an active participant.

Platform Perspective: Why the Algorithm Loves This

From a platform perspective, this video is "high-retention" content. The rapid cuts every 2-3 seconds prevent "scroll-away" boredom. Furthermore, the use of trending AI tech (Kling O1) places the video in a high-interest niche that platforms are currently promoting to compete with YouTube's educational dominance.

5 Testable Viral Hypotheses

Hypothesis 1: Side-by-side "Before/After" visuals increase save rates as users want to reference the quality later.
Hypothesis 2: Using recognizable IP (Harry Potter, Goku) triggers "fandom" engagement and shares.
Hypothesis 3: A 10-second UI walkthrough reduces the "barrier to entry" for viewers, making them feel they can do it too.
Hypothesis 4: Keeping a consistent physical anchor (the white hat) prevents the AI from looking "too fake" or disconnected.
Hypothesis 5: Ending with a physical prop (the clapperboard) makes the CTA feel more intentional and less like a generic overlay.

How to Recreate: From 0 to 1

Source Your "Base" Footage: Record yourself in a simple environment (green screen is best, but a plain wall works) performing a clear action, like holding a stick or making a gesture.
Maintain Character Consistency: Wear a distinct accessory (like the white cap in the video). This helps the AI "lock" onto your identity across different generations.
Select Your Reference Image: Find or generate a high-quality image of the style you want (e.g., a cinematic shot of a wizard's hall).
Use Kling O1 (or similar Video-to-Video tools): Upload your base video and your reference image. Use the "Transformation" or "Video Reference" mode.
Craft the Prompt: Use a prompt that describes the change. Example: "Replace the background with a snowy mountain and change the subject's clothes to a fur-lined cloak."
Generate Multiple Angles: If your tool supports it, use the "Multiple Angles" feature to create a sense of space, as shown in the bedroom scene (00:09).
Edit for Rhythm: Cut your video so the "AI reveal" happens on a beat. Use a 1:1 or 9:16 split-screen to show the comparison.
Add the Engagement CTA: Use an on-screen text overlay or a physical prop at the end to tell users exactly what to comment to get the tool link.

Growth Playbook

Opening Hook Lines

"This new AI update just killed traditional video editing."
"I turned my bedroom into a Hollywood set with one click."
"Stop spending hours on VFX. Do this instead."

Caption Templates

The "Tool Reveal" Template:
Kling’s New AI Model “O1” just released and it’s a game changer. 🤯

You can now edit lighting, clothing, and environments using just text prompts. No more complex masking.

Want to try it yourself?

👇 Comment “AI” and I’ll send you the direct link!

Hashtag Strategy

Broad (Reach): #AI #VideoEditing #ContentCreator #TechTrends
Mid-Tier (Niche): #KlingAI #AIVideo #VFX #Filmmaking
Long-Tail (Specific): #KlingO1Tutorial #AIforCreators #VideoTransformation

FAQ

What tools make it look the most similar?

Kling O1 is the specific tool used here, but Runway Gen-3 and Luma Dream Machine offer similar video-to-video features.

What are the 3 most important words in the prompt?

"Consistency," "Lighting-match," and "High-fidelity" are key to getting that cinematic look.

Why does the generated face look inconsistent?

AI often struggles with faces; using a reference image of yourself (Face Swap or IP-Adapter) helps maintain identity.

How can I avoid making it look like AI?

Keep the movement subtle and ensure the lighting on your "base" footage roughly matches the target scene.

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

Instagram currently favors high-quality "aesthetic" AI transformations, while TikTok prefers "how-to" and "chaos" AI content.