Our $1,000,000 AI film festival submission 🚨 I took 6 top Hollywood cinematographers with ZERO AI experience and taught them a simple framework to master the tools. You can easily learn this process in minutes 👇 Here are the exact tips, prompts, and framework we used: 🧵 https://t.co/Rt8PiCs2T0

PJ Ace

@PJaccetturo · creator

TWITTER · 2025-12-21Source

1.3Klikes

133comments

Remix This

Prompt

GLOBAL LOCK: Cinematic film style, high-production value, 35mm anamorphic lens feel, high contrast, dramatic lighting. Consistent male voiceover: warm, emotional, mid-range, weary tone. Color grade: Teal and orange palette, desaturated for prison, vibrant for cafe.

[00:00–00:07] 
Subject: Middle-aged man, Caucasian, short graying hair, stubble, wearing an orange prison jumpsuit. 
Environment: Dimly lit prison cell, concrete walls, harsh light coming through a small barred window. 
Action: Close-up on his face, eyes filled with tears and desperation, looking directly into the camera. 
Camera: MCU, static, shallow depth of field. 
Lighting: Low-key, side-lit, dramatic shadows. 
Speech: "I’m going to die today. In seven minutes, they’re giving me the needle."

[00:08–00:22] 
Subject: Rapid cuts of different people: a skydiver in mid-air, an old man on a park bench, a construction worker in a manhole, a pilot in a cockpit. 
Environment: Sky, NYC park, city street at night, airplane cockpit. 
Action: Skydiver screaming in exhilaration, old man watching pigeons, worker looking around confused, pilot panicking. 
Camera: POV, wide shots, handheld movement. 
Lighting: Bright daylight, sunset, harsh streetlights. 
Speech: "Every 24 hours, I wake up in another person's body. Every single day, it's a different person, a different life."

[00:23–00:40] 
Subject: A young man with curly dark hair and a young woman with short red hair (Quinn). 
Environment: A cozy, sunlit Parisian cafe, large windows, vintage decor. 
Action: They are sitting across from each other, talking and laughing. He looks at her with intense love. 
Camera: Over-the-shoulder MCU, soft panning. 
Lighting: Warm, golden hour, soft-focus background. 
Speech: "It started on the day that Quinn and I signed up for this couples' sleep study. We needed the cash."

[00:41–01:05] 
Subject: Montage: Underwater welder in a dark sea, a homeless man with a long beard laughing, a man in a cyber cafe typing frantically. 
Environment: Deep ocean, rainy city street, dark internet cafe. 
Action: Welder sparks flying, homeless man hugging himself in the rain, man staring at an "Error" message on a screen. 
Camera: ECU on eyes, dynamic tracking shots. 
Lighting: Dark, industrial, neon accents. 
Speech: "I kept waking up as... me, but not me. I had this crazy feeling like I was going insane. All I knew was that I had to find Quinn."

[01:06–01:30] 
Subject: A man in a gold patterned shirt at a luxury party, then driving a convertible sports car. 
Environment: Opulent mansion, high-speed highway at sunset. 
Action: He is dancing wildly with a drink, then laughing hysterically while driving fast. 
Camera: Wide shots, drone-like tracking, fast cuts. 
Lighting: High-key, glamorous, flashing party lights. 
Speech: "I could do anything I wanted. So I did everything. And then some. I thought there'd be no consequences."

[01:31–01:50] 
Subject: The curly-haired man waking up on a park bench in Paris, then running through the streets. 
Environment: Eiffel Tower in the background, Parisian bridge at dawn. 
Action: He wakes up, realizes where he is, and sprints toward the cafe. 
Camera: Tracking shot behind him as he runs, low angle. 
Lighting: Soft morning light, cool blue tones. 
Speech: "And after two years and 17 days, I woke up in the one place that I hadn't been able to look."

[01:51–02:10] 
Subject: He enters the cafe and finds Quinn (red hair, yellow scarf). They embrace and cry. 
Environment: The same Parisian cafe, now quieter. 
Action: They hold each other tightly, whispering. 
Camera: MCU, slow zoom-in on their faces. 
Lighting: Warm, intimate, soft rim light. 
Speech: "Jason... I knew you'd make it one day. I knew it."

[02:11–02:30] 
Subject: Back to the man in the orange jumpsuit, strapped to an execution gurney. A priest (older man) stands over him. 
Environment: Clinical execution chamber, white walls, bright overhead lights. 
Action: The man is terrified, looking at the priest. The priest looks back with a mysterious, knowing expression. 
Camera: Top-down WS, then ECU on eyes. 
Lighting: Harsh, clinical, cold. 
Speech: "Quinn... this is the end. No, Jason. This is just the beginning. Go find her, Jason."

NEGATIVE PROMPT: Cartoonish, 3D render, low resolution, blurry faces, inconsistent lighting, robotic movement, text overlays (except subtitles), flickering, morphing limbs, unnatural skin textures, flat lighting, over-saturated colors.

SPEECH PACK:
[00:00] "I'm going to die today." (TAKE_A: Somber, TAKE_B: Terrified, TAKE_C: Resigned)
[00:25] "Every 24 hours, I wake up in another person's body." (TAKE_A: Explanatory, TAKE_B: Haunted)
[01:45] "Jason... it's you." (TAKE_A: Whispered, TAKE_B: Sobbing)
[02:20] "Go find her, Jason." (TAKE_A: Authoritative, TAKE_B: Compassionate)

How PJaccetturo Made This 7 Minutes to Live AI Video and How to Recreate It

"7 Minutes to Live" is a masterclass in AI-assisted storytelling, originally submitted to a $1,000,000 AI film festival. It leverages a high-concept "body-swapping" narrative to showcase the versatility of modern AI video tools. The aesthetic is purely cinematic—think Hollywood-grade color grading, diverse global locations, and a poignant, emotional core. By blending high-octane action (skydiving, rodeo) with intimate character moments (a Parisian cafe, a prison cell), the creator demonstrates how AI can bridge the gap between "indie budget" and "blockbuster visuals." This case study explores how to use AI not just for "cool clips," but for cohesive, long-form narrative structure.

What You’re Seeing: A Visual Breakdown

The video follows a protagonist who wakes up in a different body every 24 hours for exactly seven minutes. Visually, this is achieved through a rapid-fire montage of distinct characters and environments. We see a gritty prison cell, a sun-drenched rodeo, a clinical laboratory, and a romantic cafe in Paris. The wardrobe changes with every cut—from orange jumpsuits to pilot uniforms and luxury silk robes.

The lighting is highly intentional, using low-key, high-contrast setups for dramatic scenes and warm, soft-focus "golden hour" lighting for the romantic flashbacks. The color palette shifts from desaturated teals and grays in the prison to vibrant, saturated oranges and yellows in the cafe, creating a clear emotional distinction between the protagonist's "reality" and his "memory."

Shot-by-Shot Analysis

Time Range	Visual Content	Shot Language	Lighting & Color	Viewer Intent
00:00–00:07	Man in prison cell, close-up on face.	CU, Static, Shallow DOF	Moody teal, high contrast	Hook: Immediate stakes and mystery.
00:08–00:25	Rapid montage: Skydiving, park bench, construction.	POV, Wide, Handheld feel	Naturalistic, varied	Establish the "body-swapping" mechanic.
00:26–00:45	Hospital lab, meeting Quinn in a cafe.	MCU, Over-the-shoulder	Warm, romantic, soft glow	Emotional core: Introduce the love interest.
00:46–01:10	Underwater welding, homeless man, cyber cafe.	ECU, Dynamic tracking	Dark, gritty, industrial	Rising action: The struggle to find her.
01:11–01:35	Mansion party, luxury cars, hospital emergency.	WS, Drone-like movement	High-key, glamorous vs. clinical	Climax: The chaos of living multiple lives.
01:36–02:00	Final embrace in Paris, return to prison.	MCU, Slow zoom-in	Golden hour to cold teal	Resolution: The tragic twist.

Why It Went Viral: The Narrative Hook

The primary driver of this video's success is its high-concept premise. The "7 minutes to live" hook is a classic psychological trigger—it creates an immediate sense of urgency and curiosity. Viewers aren't just watching AI-generated art; they are following a story with stakes. This taps into the biological instinct to resolve narrative tension.

Furthermore, the video capitalizes on the "AI Mastery" trend. By framing it as a submission for a million-dollar festival and mentioning that "Hollywood cinematographers" were involved, the creator adds a layer of professional prestige. This encourages saves and shares from other creators looking for "aesthetic references" or "proof of concept" for their own AI projects.

From a platform perspective, the video succeeds because of its pacing and visual variety. The algorithm rewards high retention, and the constant change in scenery and character every few seconds prevents "viewer fatigue." The emotional payoff at the end ensures that users watch until the very last frame, signaling to the platform that the content is high-value.

5 Testable Viral Hypotheses

The "High Stakes" Hook: Starting with a countdown or a life-or-death situation (0-3 seconds) increases initial retention by 40%.
Visual Contrast: Rapidly switching between extreme environments (e.g., underwater to a desert) triggers a "novelty response" in the brain, keeping the viewer engaged.
The "AI Reveal" Meta-Narrative: Content that explicitly states it was made with AI (especially if it looks high-quality) attracts a niche but highly engaged audience of tech enthusiasts and creators.
Emotional Anchor: Even in a sci-fi setting, a simple human connection (the search for "Quinn") provides the emotional weight needed for shares and saves.
Looping Narrative: Ending the video where it began (the prison cell) creates a psychological "loop" that encourages re-watching to catch missed details.

How to Recreate: From Script to Screen

1. Scripting for AI Constraints

Write a script that embraces visual variety. Since AI is currently better at short, disconnected clips than long, continuous takes, a "montage" or "body-swapping" narrative is the perfect way to hide the limitations of the technology.

2. Character Consistency Framework

While the physical bodies change, the "soul" remains. Use a consistent voiceover (ElevenLabs) and a recurring visual motif (like the yellow scarf or a specific piece of jewelry) to anchor the viewer across different scenes.

3. Generating the Visual Assets

Use Midjourney for high-fidelity character and environment references. Use these as "Image Prompts" in tools like Runway Gen-3 or Luma Dream Machine to ensure the lighting and color grade remain consistent across shots.

4. Mastering the "Cinematic" Prompt

Don't just describe the scene; describe the camera. Use terms like "35mm anamorphic lens," "shallow depth of field," "teal and orange color grade," and "slow dolly zoom" to get that professional film look.

5. Sound Design is 50% of the Movie

The emotional impact of this video comes from the BGM and the voiceover. Use a cinematic orchestral track that builds in intensity and a voiceover that sounds intimate and weary.

6. Editing for Pacing

Cut on the beat. In the montage sections, shots should rarely last longer than 1.5 seconds. In the emotional scenes, let the camera linger for 3-4 seconds to allow the viewer to connect with the character.

7. Color Grading for Cohesion

Use a global LUT (Look Up Table) in your editing software (Premiere or DaVinci) to tie all the disparate AI clips together. This is the "secret sauce" that makes AI clips look like a real movie.

8. The "Hook" Thumbnail and Title

Your title should pose a question or a challenge. "7 Minutes to Live" is better than "My AI Short Film." The thumbnail should be a high-contrast close-up of a human face showing intense emotion.

Growth Playbook: Distribution Strategy

Opening Hook Lines

"I have 7 minutes to live, and I've spent every one of them looking for you."
"What would you do if you woke up in a different body every single day?"
"We taught Hollywood pros how to use AI, and this was the result."

Caption Templates

The Storyteller:
[Hook: The clock is ticking.] I woke up in a prison cell today, but yesterday I was in Paris. This is the story of a man lost in time. [Value: Created entirely with AI tools.] What would you do with your last 7 minutes? 👇 #AIFilm #Storytelling

Hashtag Strategy

Broad: #AI #Filmmaking #ShortFilm #Cinematography
Mid-tier: #RunwayML #LumaAI #MidjourneyArt #AIVideo
Niche: #AIFilmFestival #IndieCreator #VisualStorytelling

Frequently Asked Questions

What tools make it look the most similar?

Use Runway Gen-3 for video, Midjourney for style references, and ElevenLabs for the voiceover.

What are the 3 most important words in the prompt?

"Anamorphic," "Cinematic," and "High-Contrast."

How do I keep the character's face consistent?

Use the same "Seed" number and a strong "Image Prompt" of the character's face in every generation.

Why does my AI video look like a cartoon?

Add "photorealistic," "raw film stock," and "8k resolution" to your negative or positive prompts.

Is it easier to go viral on Instagram or TikTok?

TikTok favors the narrative hook, while Instagram favors the high-end cinematic aesthetic.