0:00 / 0:00

▶

Kelly Boesch

@kelly_boesch_ai_art

INSTAGRAM · 2026-02-21Source

5.9Klikes

135comments

Remix This

Recreate with Veo3.1

Make your own AI viral video

Prompt

GLOBAL LOCK: A series of surreal editorial caricature characters. All characters have exaggerated features: very large eyes, long thin necks, and sculptural, gravity-defying hair. The style is hyper-realistic but with distorted proportions. The environment is a mix of retro-surreal studios and vintage locations. Lighting is dramatic, high-fashion editorial style with frequent camera flashes. Color grade is muted with specific pops of orange and teal. Pacing is fast, synced to camera shutter clicks.

[00:00–00:07]
Subject: An elderly woman with pale skin, wearing large round black glasses and an orange-and-brown patterned shirt. Her hair is styled into two massive, tight grey-and-white swirls on either side of her head.
Action: She poses nervously, hands clasped, while multiple hands hold cameras and point at her. A camera flash goes off repeatedly.
Camera: Medium close-up, static.
Lighting: Soft key light with blinding white strobe flashes.

[00:07–00:15]
Subject: A thin, pale character with wild, frizzy white hair and round sunglasses, wearing a blue pinstripe suit and bowtie.
Action: Sitting in a large orange egg-shaped chair, smoking a cigarette. Thick white smoke billows out and fills the frame. Photographers are visible in the background.
Camera: Wide shot, side profile.
Lighting: Moody side-lighting, smoke catching the light.

[00:15–00:22]
Subject: A woman with dark blue hair styled in a massive, horizontal swirl. She wears a red dress with large white polka dots and round sunglasses.
Action: She sits at a table, looking annoyed, as a bald photographer with a vintage camera takes her picture.
Camera: Medium shot, slight zoom in.
Lighting: Bright studio lighting, flat but high-quality.

[00:37–00:45]
Subject: A character with a massive, vertical grey hair bun wearing large headphones.
Action: Holding a saucer and cup, taking a slow sip of steaming tea. A photographer in a brown coat is visible to the left.
Camera: Medium shot, profile view.
Lighting: Soft, motivated by a nearby studio lamp.

[01:04–01:12]
Subject: A man with a thin face, large round glasses, and a small black cap. His hair is styled in two white puffs.
Action: He looks directly into the lens with a worried expression. Multiple cameras surround him, and rapid-fire flashes explode, turning the screen white momentarily.
Camera: Extreme close-up on the face.
Lighting: Intense, rhythmic strobe flashes.

[01:28–01:36]
Subject: An elderly man with wild grey hair and a tiny black top hat, wearing a pinstripe suit.
Action: Sitting in a vintage teal train car, reading a large book. A woman in a striped dress takes his photo from the side.
Camera: Medium shot, interior train setting.
Lighting: Warm, cinematic interior lighting.

[01:51–01:57]
Subject: A woman with a perfectly round grey hair bun and round sunglasses.
Action: Sitting in a chair in a cluttered room. A large black crow perches on a vintage TV next to her.
Camera: Medium wide shot.
Lighting: Dim, moody, "lived-in" feel.

[02:35–02:43]
Subject: A woman with massive, cloud-like blonde hair and large round glasses.
Action: She looks distressed as two assistants or photographers adjust her and take photos in a dark, moody room.
Camera: Medium shot, centered.
Lighting: Low-key, dramatic, fading to black.

NEGATIVE PROMPT: generic 3D render, low resolution, blurry textures, normal human proportions, messy hair (unless specified), flat lighting, no camera flashes, modern digital camera look, inconsistent character features, robotic movement, bad anatomy.

SPEECH PACK:
No spoken dialogue.
Audio consists of:
- Upbeat, rhythmic electronic BGM.
- Frequent, sharp "click-clack" sounds of mechanical camera shutters.
- High-pitched "whirring" and "pop" sounds of studio flashes charging and firing.
- Occasional soft vocalizations: [sigh] at 00:54, [gasp] at 01:32.

TAKE_A: Focus on the rhythmic timing of the shutter clicks.
TAKE_B: Emphasize the atmospheric room tone in the quieter moments.
TAKE_C: Maximize the intensity of the flash sounds during the tunnel sequence.

How kelly_boesch_ai_art Made This Surreal Editorial Caricature AI Video Guide — and How to Recreate It

This case study analyzes a viral AI-generated video featuring a series of surreal editorial caricatures. The aesthetic blends high-fashion photography with grotesque, exaggerated character designs—think Wes Anderson meets Tim Burton in a high-end fashion magazine. Key visual elements include sculptural, gravity-defying hairstyles, hyper-detailed skin textures, and a "behind-the-scenes" narrative where these bizarre subjects are hounded by photographers. The video leverages a rhythmic, camera-shutter-synced editing style that creates a hypnotic viewing experience, perfect for Instagram and TikTok's fast-paced consumption.

What You’re Seeing

The video is a montage of distinct "fashion shoots." Each segment introduces a new, highly stylized character—often elderly, with elongated limbs, massive eyes, and hair styled into impossible swirls, buns, or wild clouds. The subjects are dressed in retro-chic attire (polka dots, pinstripes, turtlenecks) and placed in minimalist, slightly surreal environments like a wooden studio, a vintage train car, or a concrete bunker. The primary action involves the characters posing, smoking, or drinking tea while being bombarded by camera flashes and clicking shutters.

Shot-by-Shot Breakdown

Time Range	Visual Content	Shot Language	Lighting & Color	Viewer Intent
00:00–00:07	Elderly woman with swirl hair in orange patterns.	Medium Close-up (MCU), static.	Soft studio light, muted teal background.	Hook: Establish the "weird editorial" world.
00:07–00:15	Character in an orange egg chair smoking.	Wide Shot (WS), side profile.	Dramatic side-lighting, heavy smoke.	Atmosphere: Create a sense of "cool" absurdity.
00:29–00:37	Character with wild white hair checking a phone.	Medium Shot (MS), low angle.	High contrast, cool tones.	Relatability: Modern tech in a surreal setting.
01:04–01:12	Character in a tunnel with rapid camera flashes.	Close-up (CU), centered.	Strobe lighting, blinding white flashes.	Sensory Overload: Increase intensity and engagement.
01:28–01:36	Character in a train car reading a book.	Medium Shot (MS), tracking feel.	Warm interior light, teal train walls.	Narrative: Show the character in a "real" world.

Why It Went Viral

The Psychology of the "Uncanny Editorial"

This video taps into the "Uncanny Valley" effect but steers it toward artistic expression rather than horror. By using caricatures, the creator avoids the "creepy" factor of near-human AI and instead creates a distinctive brand identity. The subject matter—the absurdity of high-fashion photography—is a globally recognized trope. It mocks the self-importance of the industry while simultaneously being visually stunning, satisfying both the viewer's desire for humor and their appreciation for high-fidelity art.

Platform Synergy & Pacing

From a platform perspective, the video is a masterclass in retention-based editing. The 0–3 second hook is immediate: a bizarre face you can't look away from. The use of diegetic sound effects (the camera shutters) synced perfectly to the music creates a rhythmic "ASMR" effect. This encourages users to watch until the end to see the next character reveal. The high "save" rate likely comes from other creators and artists using it as a style reference or "mood board" inspiration.

5 Testable Viral Hypotheses

The Rhythmic Shutter Effect: Syncing visual cuts to the sound of a camera shutter increases watch time by creating a predictable but satisfying pattern.
Exaggerated Character Consistency: Using AI to maintain a specific "caricature" style across different scenes builds a stronger brand than generic "cinematic" AI.
The "Behind-the-Scenes" Meta-Narrative: Showing the "photographers" in the frame makes the AI generation feel like a real production, increasing perceived value.

High-Contrast Lighting Hooks:

Retro-Futuristic Color Palettes: Combining 1960s fashion with modern AI textures creates a "timeless" look that appeals to a broad demographic.

How to Recreate (Step-by-Step)

Define Your "Core Weirdness": Create a character sheet. Are they elongated? Do they have swirl hair? Use Midjourney to generate 5-10 consistent reference images of one character.
Master the "Editorial" Prompt: Use keywords like "editorial fashion photography," "high-end magazine style," "dramatic rim lighting," and "Kodak Portra 400" to get that professional look.
Generate Base Frames: Use your reference images to generate high-quality stills in Midjourney or DALL-E 3. Ensure the "photographer" or "camera" is part of the prompt for some shots.
Animate with Motion Control: Use Luma Dream Machine or Runway Gen-3. Use "Motion Brush" to specifically animate the hair blowing or the character taking a sip of tea.
The "Flash" Technique: In your video generator, add a prompt for "bright camera flash lighting" to create those high-energy transitions.
Sound Design is Key: Don't just use a song. Layer in high-quality "camera shutter click" and "flash charging" sound effects.
Color Grade for Cohesion: Bring all clips into CapCut or Premiere. Apply a consistent LUT (Look Up Table) to ensure the teal and orange tones match across all AI generations.
The "Loop" Edit: Ensure the last shot has a similar composition to the first shot to create a seamless loop for Instagram Reels.

Growth Playbook

3 Opening Hook Lines

"The fashion industry is getting weirder... and I love it."
"POV: You're the most famous person in a world that doesn't exist."
"AI isn't just for realism; it's for the surreal."

Caption Templates

The "Artist's Note" Style:
Exploring the boundary between caricature and high fashion. 📸 Which character is your favorite? Let me know in the comments! 👇 #AIArt #Surrealism #FashionPhotography

The "Behind the Scenes" Style:
The paparazzi won't leave them alone. 💅 A deep dive into my latest AI character series. Created with Midjourney + Luma. #CreativeAI #DigitalArt #Editorial

Hashtag Strategy

Broad: #AIArt #DigitalArtist #Surrealism #Fashion (High reach)
Mid-tier: #MidjourneyArt #LumaAI #EditorialPhotography #CharacterDesign (Targeted)
Niche: #AIFashion #SurrealFashion #CaricatureArt #KellyBoeschStyle (Community focus)

FAQ

What tools make it look the most similar?

Midjourney for the base images and Luma Dream Machine for the high-fidelity motion.

What are the 3 most important words in the prompt?

"Editorial," "Caricature," and "Cinematic Lighting."

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 generic AI?

Add specific "imperfections" like film grain, lens flares, and exaggerated anatomical proportions.

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

Instagram, as the "aesthetic" and "fashion" communities are more established there.