Pre-Vis at Lightspeed: A Director’s Guide to AI-Powered Storyboarding
Is AI coming for your director’s chair? The answer is an emphatic no. But a filmmaker, a cinematographer, or a game designer who knows how to command an army of AI concept artists will run laps around one who doesn’t. As of July 8, 2025, the lengthy and astronomically expensive process of visual development has been fundamentally ruptured. Forget the panicked headlines. Today, we aren’t just talking theory; we are opening up the Creative Lab and turning Midjourney into a world-class, on-demand pre-visualization department for your next big idea.
For decades, translating the grand vision in your head to a tangible visual language has been one of the biggest hurdles in creation. You sketch, you hire a concept artist, you build mood boards, you wait. This process is slow, expensive, and often a point of friction. What if you could generate a dozen cinematic keyframes for your opening scene before you’ve even finished your morning coffee? What if you could A/B test a Giallo horror color grade against a gritty David Fincher-esque palette in minutes, not days?
This isn’t science fiction anymore. This is the new workflow. We’re moving from a paradigm of manual creation to one of creative direction. Your role shifts from the lone painter to the conductor of an infinite orchestra of visual ideas. The goal isn’t to have AI ‘make the movie’ for you. The goal is to use AI to explore possibilities at a scale and speed that was previously unimaginable, allowing you to arrive at a stronger, more refined vision, faster.
The AI Pre-Vis Lab: A Sci-Fi Epic in Sixty Minutes
Let’s get our hands dirty. To demonstrate this power, we’re going to tackle a classic sci-fi pitch. We’ll be the director, production designer, and cinematographer all at once, using nothing but text prompts. Our mission: develop the key visual beats for a scene from an imagined film.
The Logline: In a newly terraformed crater on Mars, a lone botanist on a routine survey discovers a cave system containing a self-sustaining, bioluminescent forest that defies all known biology.
Our task is to create the core visuals for this moment of discovery: the establishing shot, the character’s reaction, and the details that make the world feel real. Let’s fire up our creative co-pilot.
Part 1: The Establishing Shot – Your Cinematic Seed
Every great scene needs a powerful establishing shot. This is the image that sets the tone, scale, and mood. We don’t want a generic ‘space forest.’ We want a specific feeling. We need to tell the AI not just what to draw, but how to ‘shoot’ it. This is where your latent knowledge as a creator truly shines.
The Prompting Studio: The Establishing Shot
Fire up Discord and head to your Midjourney channel. We are going to feed the AI a rich, cinematic recipe.
Copy and paste this master prompt:
/imagine prompt: cinematic film still from a Denis Villeneuve sci-fi epic, a lone astronaut botanist stands at the entrance of a vast cave on Mars, inside is a shimmering bioluminescent forest of glowing fungi and otherworldly flora, ethereal god rays filtering through the cavern ceiling, atmospheric haze, wide shot, shot on Arri Alexa 65, anamorphic lens flare –ar 21:9 –s 250 –style raw –v 6.0
Hit Enter. While Midjourney works its magic for the next 60 seconds, let’s break down exactly what we just did and why it’s so powerful.
Strategist’s Log (Deconstructing the Shot): This prompt is a miniature film production plan. We specified a director (‘Denis Villeneuve’) to give the AI a strong stylistic anchor—think Blade Runner 2049 or Dune. We mentioned specific camera gear (‘Arri Alexa 65’) and lenses (‘anamorphic’) to evoke high-end cinematography. Terms like ‘atmospheric haze’ and ‘god rays’ are art direction notes. This isn’t just asking for a picture; it’s giving creative briefs.
Strategist’s Log (Decoding the Parameters): The text is only half the story. The parameters are your technical controls. –ar 21:9 forces the ultra-widescreen ‘CinemaScope’ aspect ratio, immediately making it feel more like a film. –s 250 (stylize) tells Midjourney to follow our stylistic instructions with high fidelity. –style raw gives us a more photographic, less ‘opinionated’ AI base image, which is great for realism. Finally, `–v 6.0` ensures we’re using the latest, most powerful version of the model.
Within a minute, Midjourney will present you with a grid of four stunningly different interpretations of your command. This is your first major creative decision point. You are no longer just an ideator; you are a curator. Which of these four concepts best captures the soul of your scene? Select one to ‘upscale’—that becomes your ‘hero’ image, your cinematic seed.
Part 2: Building the Scene – Directing the AI’s Camera
An establishing shot is not a scene. Now we need coverage. In traditional filmmaking, this would mean moving the camera, changing lenses, and getting new shots. With Midjourney, we can do the digital equivalent using its powerful variation tools, specifically Vary (Region).
Once you’ve upscaled your hero image, you’ll see a ‘Vary (Region)’ button. Clicking this opens an editor where you can select a part of your image and rewrite the prompt *just for that area*. This is a paradigm-shifting tool for creators. You are no longer re-rolling the entire image; you are giving notes on a specific part of the frame.
The Prompting Studio: The Reaction Shot
Click ‘Vary (Region)’ on your upscaled establishing shot. Use the lasso tool to carefully select the head and shoulders of your astronaut. Now, we’re going to punch in for a close-up.
In the prompt box that appears, replace the original prompt with this:
/imagine prompt: cinematic medium close-up of the astronaut’s face, helmet visor showing a crystal clear reflection of the glowing forest, a look of profound awe and disbelief, subtle anamorphic lens flare –ar 21:9 –style raw
Midjourney will now regenerate only the part you selected, seamlessly blending it with the unchanged background. You’ve just directed your AI cinematographer to get a new shot without moving the lights.
The result is a perfect reaction shot that maintains the lighting, color, and atmosphere of your original wide shot. You can repeat this process indefinitely. Select a glowing mushroom on the ground and prompt for an ‘extreme close-up macro shot showing detailed alien texture’. Select the cave ceiling and prompt for ‘light filtering through crystal formations’. In under ten minutes, you can build a complete, visually cohesive storyboard sequence that would have taken a concept artist days to illustrate.
Part 3: Maintaining Visual Cohesion – Your Style Bible
The key to making a series of AI generations feel like a single, unified project is creating a ‘Style Bible’—a core set of keywords that you include in every prompt for the project. This acts as a creative fingerprint.
For our Mars film, the style bible might be:
cinematic film still, Denis Villeneuve style, Arri Alexa 65, anamorphic lens, atmospheric haze --ar 21:9 --style raw
By keeping this block of text consistent, you can change the ‘subject’ of the prompt while maintaining the ‘look’.
- For a medium shot: `[STYLE BIBLE] medium shot of the botanist cautiously reaching out to touch a glowing plant…`
- For an insert shot: `[STYLE BIBLE] insert shot of the astronaut’s gloved hand hovering inches away from a pulsating blue fungus…`
- For a point-of-view shot: `[STYLE BIBLE] POV shot from inside the helmet, looking out at the dazzling alien landscape…`
This method ensures every image you generate feels like it was shot on the same day, with the same crew, on the same location. It elevates a collection of cool images into a believable narrative sequence.
Once you have your keyframes—your establishing shot, medium, close-up, and insert—you can lay them out in a simple storyboard format using any presentation software. What you’re looking at is no longer a set of disparate AI images. It’s pre-production. It’s a visual treatment you can use to pitch producers, brief your VFX team, or simply solidify your own directorial vision.
The Big Questions: Your AI Debrief
“Is this legal for commercial work? What about copyright?”
This is the most critical question. The landscape is evolving, but here’s the current consensus for best practice. AI models like Midjourney have terms of service that grant you broad commercial rights to the images you create, especially with a paid plan. However, the legal concept of ‘copyright’ for purely AI-generated images is still being debated globally. The smartest workflow is to use AI output as a foundational element. Generate your concept art, then bring it into Photoshop, paint over it, composite it, add custom graphics, and significantly alter it. By adding a substantial layer of human creativity, you are transforming the work and strengthening your claim to authorship. For pre-visualization and concepting, it’s an unparalleled tool. For final assets, always assume a ‘human touch’ phase is required.
“How do I avoid my work looking generic and ‘AI-ish’?”
The ‘generic AI look’ comes from generic prompts. The antidote is specificity and a multi-stage workflow. Never use the first output as your final product. First, get hyper-specific with your prompt. Don’t say ‘a knight’; say ‘a weathered knight in tarnished, gothic-style plate armor, inspired by the art of Zdzisław Beksiński.’ Second, iterate relentlessly. Use the ‘Vary’ and ‘Remix’ tools to explore tangents. Third, and most importantly, perform post-production. Take the 90% solution from the AI into a human-centric tool like Photoshop, Affinity Photo, or DaVinci Resolve. Tweak the color grade, add film grain, composite multiple AI generations together, and paint in your own unique details. The AI is your first draft; your artistry provides the final polish that makes it unique.
“Will this replace concept artists and cinematographers?”
No, it will change their roles and empower them. A concept artist who can generate fifty initial ideas in an afternoon is infinitely more valuable than one who can produce two. Their skill will shift from pure illustration to ideation, art direction, and masterful refinement of AI-generated bases. Similarly, a cinematographer can use these tools to have a more concrete visual conversation with the director long before they ever get on set. Imagine being able to pre-visualize three different lighting setups for a scene and share them with the gaffer. It’s a tool for communication and exploration that enhances, not replaces, the craft.
Your Creative Sandbox Assignment
You’ve seen the workflow. Now it’s your turn to get in the director’s chair. Your mission is to create a 3-shot storyboard for a completely different genre and setting.
Your Logline: A hardboiled detective in a rain-soaked, 1940s noir version of New York City finds a mysterious, glowing artifact in a smoky jazz club.
Your task is to use Midjourney to create:
- The Establishing Shot: A wide shot of the noir city street, rain slicking the cobblestones, neon signs reflecting in the puddles.
- The Medium Shot: The detective, trench coat collar up, sitting at the bar in a smoky, dimly-lit jazz club.
- The Insert Shot: A close-up of the glowing artifact resting on the mahogany bar top, casting an unnatural light.
Focus on creating your own ‘Style Bible’. What keywords will you use? ‘Film noir,’ ‘chiaroscuro lighting,’ ‘shot on black and white Kodak Double-X film,’ ‘deep shadows,’ ‘hardboiled detective’? Experiment and see how you can direct the AI to your specific vision.
Your AI Integration Plan This Week
Embracing this new toolset requires practice. Here’s a simple, manageable plan to integrate this workflow into your creative habit this week.
- Monday: Idea Day. Forget the AI. Just write down three single-sentence loglines for scenes you find exciting. A pirate ship in a nebula? A medieval kitchen during a magical feast? A spy handoff on a foggy bridge?
- Wednesday: The Establishing Shot. Pick your favorite logline from Monday. Spend 20-30 minutes in Midjourney focusing *only* on creating the perfect establishing shot. Don’t worry about other shots. Just nail the main image. Iterate on your prompt until the mood feels right.
- Friday: Scene Expansion. Take your favorite image from Wednesday. Spend 30 minutes using Vary (Region) and your established ‘Style Bible’ to generate two more shots: a medium close-up of a key character or object, and an interesting detail shot.
- Sunday: The Review. Drop your three images into a Google Slide or a Keynote presentation. Look at them in sequence. Did you successfully create a micro-narrative? What did you learn about prompting? This simple loop is the key to mastering AI as a creative partner.
We are at the very beginning of a seismic shift in creative production. The tools we’ve explored today are not endpoints, but starting blocks. They are fluency multipliers. They collapse the time between idea and execution, allowing you to live in a state of constant visual exploration. The fear of being replaced by AI is misplaced. The real opportunity is to augment your unique, human vision with a co-pilot that never runs out of ink, film, or inspiration. Now go make something incredible.
Post Comment
You must be logged in to post a comment.