0:00 / 0:00

Adriana Bubori

@adrianabubori

INSTAGRAM · 2026-02-12Source

498likes

313comments

Remix This

Recreate with Kling 3

Make your own AI viral video

Prompt

GLOBAL LOCK:
Subject 1: A woman with dark brown skin, warm undertones, approximately 25 years old. She has intricate dark braids (cornrows) pulled back. She wears large, chunky gold hoop earrings and a white top. Her makeup features a bold, matte red lipstick and groomed eyebrows.
Subject 2: A woman with light-medium skin tone, approximately 28 years old, with wavy chestnut brown hair. She wears pink and white vertical striped silk pajamas.
Environment: Shot 1-4 in a bright, clean studio setting with soft diffused lighting. Shot 5-9 in an outdoor garden setting with soft golden hour sunlight and lush green bokeh background.
Style: High-end fashion editorial photography, hyper-realistic, 8k resolution, extreme focus on skin texture (pores, fine lines, freckles), iPhone 15 Pro cinematic aesthetic.
Audio/Speech: No speech, but rhythmic percussive beats drive the cuts.

[00:00–00:01]
Subject 1 in a Medium Close-Up (MCU). She rests her chin on her hand, looking directly into the camera with a neutral, confident expression. Soft studio lighting from the front-left. Camera is static.

[00:01–00:02]
Extreme Close-Up (ECU) of Subject 1's lips. The matte red lipstick shows realistic lip texture and fine lines. The skin around the mouth shows visible pores and slight peach fuzz. Teeth are slightly visible behind the lips. Static macro shot.

[00:02–00:03]
Extreme Close-Up (ECU) of Subject 1's right eye. Dark brown iris with a clear catchlight. Individual eyelashes are distinct. Skin texture around the eye shows natural folds and fine lines. Static macro shot.

[00:03–00:04]
Extreme Close-Up (ECU) of Subject 1's nose and cheek area. Focus on realistic freckles and skin pores. Side-lighting emphasizes the texture of the skin. Static macro shot.

[00:04–00:05]
Subject 2 in a Medium Close-Up (MCU). She is outdoors, smiling warmly at the camera. Her wavy hair catches the sunlight. Background is a soft green blur. Natural, warm lighting. Camera is static.

[00:05–00:06]
Extreme Close-Up (ECU) of Subject 2's eye. Green/hazel iris. Visible fine lines (crow's feet) at the corner of the eye, showing realistic aging/expression. Static macro shot.

[00:06–00:07]
Extreme Close-Up (ECU) of Subject 2's mouth as she smiles. Natural white teeth with slight realistic imperfections. Skin texture around the smile lines is highly detailed. Static macro shot.

[00:07–00:08]
Close-Up (CU) of the pink striped pajama pocket. The word "editz" is embroidered in matching pink thread. The weave of the fabric and the texture of the embroidery thread are clearly visible. Static macro shot.

[00:08–00:09]
Close-Up (CU) of the pajama buttons. Two small, pearlescent buttons on the pink striped fabric. Focus on the material detail and the stitching. Static macro shot.

NEGATIVE PROMPT:
Smooth waxy skin, blurry textures, distorted eyes, unrealistic teeth, plastic-looking fabric, flickering light, morphing features, low resolution, cartoonish colors, robotic movement, floating hair, missing pores, airbrushed look.

SPEECH PACK:
(No speech present in video. The video relies on visual texture and rhythmic editing.)

How adrianabubori Made This Hyper Realistic AI Skin Texture Editorial Tutorial — and How to Recreate It

This case study examines a hyper-realistic AI-generated video that challenges the viewer's perception of "AI-ness." The video features high-end cinematic editorial portraits with an extreme focus on macro skin texture, iPhone-style aesthetic, and warm, natural lighting. By showcasing details that AI traditionally struggles with—such as individual skin pores, fine facial hair, realistic dental imperfections, and intricate fabric embroidery—the creator @adrianabubori leverages the "uncanny valley" to spark engagement. The core hook, "I can spot AI from a mile away," serves as a direct challenge to the audience, inviting them to scrutinize the visuals and ultimately proving that high-fidelity AI can now bypass the typical "AI look."

What You’re Seeing

The video is a rhythmic montage of two distinct subjects. The first is a woman with dark braided hair and bold red lipstick in a studio-like setting. The second is a woman with wavy brown hair in pink striped pajamas, filmed in a soft, outdoor natural light environment. The editing is fast-paced, synchronized to a percussive beat, alternating between medium-close-up (MCU) shots and extreme-close-up (ECU) macro shots.

Shot-by-shot Breakdown (Estimated)

Time Range	Visual Content	Shot Language	Lighting & Color	Viewer Intent
00:00–00:01	Subject 1: Braided hair, red lips, gold hoops.	MCU, Static	Soft studio light, warm tones.	Establish the "human" subject.
00:01–00:02	Macro of red lips and skin pores.	ECU, Macro	High contrast, saturated red.	Hook: Show impossible detail.
00:02–00:03	Macro of a brown eye and eyelashes.	ECU, Macro	Warm, catchlight in eye.	Reinforce realism/texture.
00:03–00:04	Macro of nose and cheek freckles.	ECU, Macro	Side-lit to show skin relief.	Challenge the "smooth AI" trope.
00:04–00:05	Subject 2: Smiling in pink pajamas.	MCU, Outdoor	Golden hour, soft green bokeh.	Contrast with studio setting.
00:05–00:06	Macro of a green eye with fine lines.	ECU, Macro	Natural daylight.	Show age-appropriate realism.
00:06–00:07	Macro of a natural smile and teeth.	ECU, Macro	Bright, clean.	Humanize the AI generation.
00:07–00:08	Macro of "editz" embroidery on fabric.	CU, Macro	Soft, diffused.	Showcase material consistency.

Why It Went Viral

The "Challenge" Psychology

The choice of the text overlay, "I can spot AI from a mile away," is a masterclass in psychological baiting. It targets the common human desire to be "right" or "perceptive." By presenting a statement that the visuals immediately contradict, the creator forces the viewer to stop scrolling and look closer. This increases dwell time significantly, a key metric for Instagram and TikTok algorithms. The audience isn't just watching; they are investigating.

The Aesthetic of "Anti-AI"

Most AI videos are criticized for being too smooth, "waxy," or having "dream-like" movements. This video leans into the opposite: hyper-texture. By focusing on pores, peach fuzz, and the weave of a pajama pocket, it addresses the specific "tells" people look for. This "Anti-AI" aesthetic feels premium and editorial, making it shareable as an "artistic" or "tech-demo" piece.

Platform Signal Analysis

From a platform perspective, the high save-to-view ratio likely propelled this video. Creators save this as a reference for "how good AI can look," while casual users save it to show friends. The rhythmic editing (cutting on the beat) ensures high completion rates, as the visual "payoff" (the next macro shot) happens every 0.8 seconds.

5 Testable Viral Hypotheses

The Texture Trap: If you show extreme macro details (pores, fabric), viewers will watch longer to find "glitches."
The Contradictory Hook: Using a caption that denies what the video is (e.g., "This isn't AI") triggers a "fact-checking" instinct in users.
Rhythmic Pacing: Syncing hard cuts to a percussive beat creates a hypnotic effect that prevents scrolling.
Material Consistency: Showing a specific brand or word (like "editz") embroidered on fabric proves high-level control, increasing "wow" factor.
The "Human Flaw" Factor: Including "imperfections" like fine lines or slightly uneven teeth makes AI content more relatable and less "uncanny."

How to Recreate (Step-by-Step)

Topic Selection: Choose a "High-Fidelity Challenge" theme. This suits accounts focused on AI art, tech, or high-end fashion.
Character Consistency: Use a tool like Midjourney with the --cref (Character Reference) tag or Flux.1 [dev] with a LoRA to ensure the face remains identical across MCU and ECU shots.
Texture-First Prompting: When generating images, use keywords like "macro photography," "8k skin texture," "visible pores," and "fujifilm x-t4."
Keyframe Generation: Generate 8-10 high-res images: 2 MCU shots and 6-8 ECU shots of specific details (eyes, lips, fabric).
Video Generation: Use Kling AI or Luma Dream Machine. For the MCU shots, use a "subtle smile" or "gentle breathing" prompt. For the macro shots, keep motion to a minimum (0.5 intensity) to avoid warping the texture.
The "Editz" Detail: Use an AI image editor or In-painting to add specific text or logos to clothing to prove control.
Editing: Import clips into CapCut. Use a "Phonk" or "Percussive" beat. Use the "Auto-beat" feature to mark cuts.
Overlay & Publishing: Add the "I can spot AI..." text in a clean, sans-serif font (like Montserrat). Export in 4K, 30fps.

Growth Playbook

Opening Hook Lines

"They said AI couldn't do texture. They were wrong."
"Can you find the glitch in this video?"
"Stop scrolling: Is this a real person or a prompt?"

Caption Templates

The "Challenge" Template:
I bet you can't find the AI tell in this one. 🔍 We’ve reached a point where pixels feel like skin. Which shot fooled you the most? Let me know in the comments! 👇 #AIArt #Hyperrealism

The "Process" Template:
How I achieved this 8K skin texture using [Tool Name]. 🛠️ It’s all about the macro prompting and lighting consistency. Check the link in bio for my prompt guide! 🚀 #AICreator #DigitalArt

Hashtag Strategy

Broad: #AI #DigitalArt #Photography #Tech (High reach)
Mid-tier: #AIModel #MidjourneyArt #CinematicVideo #AIVideo (Targeted)
Niche: #SkinTexture #MacroPhotography #FluxAI #AIChallenge (High engagement)

FAQ

What tools make it look the most similar?

Flux.1 [dev] for the base images and Kling AI for the high-fidelity motion.

What are the 3 most important words in the prompt?

"Macro," "Pores," and "Cinematic Lighting."

Why does the generated face look inconsistent?

You aren't using a Character Reference (cref) or a consistent seed across shots.

How can I avoid making it look like AI?

Add "imperfections" like stray hairs, skin redness, or non-perfect teeth.

Is it easier to go viral on Instagram or TikTok?

Instagram, as the "aesthetic" and "editorial" vibe performs better there.