Recommended resources
The freely available resources this lesson will direct you to.
Hello again. Last lesson asked whether a generated clip helps the audience follow a story, rather than merely showing what an AI tool can do. This lesson brings that judgment into planning: how small must a first film be to finish it on a phone without spending money?
You’ll set four kinds of limits—scope, time, device, and cash—and see how they affect one another. Allow about 35–40 minutes for the lesson and its short viewing and reading selections. We are not choosing specific AI services yet; their free allowances and terms change, and you will evaluate them later in the course.
Begin with resources, not spectacle
“Zero budget” means no new cash expenditure, not unlimited production capacity. You may already have a phone and internet access, but your time, storage, battery, and any free generation allowance are finite. A story requiring six locations does not become easy simply because each location can be generated: each one still needs a recognizable look, usable shots, and continuity across the edit.
In How to Make a Short Film with No Money, Simon Horrocks describes writing around resources that are actually available. His examples concern filming with a camera, not generating images, but the planning principle transfers: decide what you can reliably make before committing to what the script demands.
How to Make a Short Film with No Money - Full Masterclass
Watch the selected part of “How to Make a Short Film with No Money” by Simon Horrocks for its approach to controlling scope before production.
Start with available resources, where Horrocks argues that a zero-budget film begins with the story. Then watch location limits for his explanation of why even free locations consume time and why one setting can be a sensible first constraint. Apply the principle to generated locations, rather than treating his live-action advice as an AI-tool recommendation.
Consider a possible micro-short: someone sees an empty chair, carries over a second cup, hesitates, and sets the cup down beside it. The film could imply absence, hope, or habit without dialogue. It has one visible character, one room, and two important props. That does not guarantee an effective film. It does give you a contained storytelling problem: make the hesitation and final choice readable.
For a first zero-cash, phone-first AI film, a useful starting ceiling is 30–45 seconds, one setting, at most one visible character, and roughly four to six planned shots. These are planning limits, not artistic rules. A second character means another appearance to maintain; another room means another environment to establish; dialogue adds voice, timing, and intelligibility work. Add any of them when the story needs them and you have verified that your resources can support them.
Set a shot limit that can survive failed generations
The finished duration is only part of the workload. Five shots do not necessarily mean five generations: an image may misplace a prop, a video may distort a hand, or a good-looking take may not cut cleanly into the next shot. The question is not “Can a tool make one impressive clip?” but “Can my available allowance produce all the moments this film requires, with room for some failures?”
The short selections below come from Standard Story Company’s How to shoot a short film... on iPhone. It is a live-action example, so do not borrow its production times as AI-generation estimates. Notice instead how the creator writes for what is available and how solo work takes longer than the initial plan.
How to shoot a short film... on iPhone
Watch these excerpts from “How to shoot a short film... on iPhone” by Standard Story Company for two planning lessons: build around available resources, and leave time for the work of doing multiple jobs yourself.
Watch available assets for the script’s starting constraint. Later, watch solo timing for the account of work extending beyond the planned hours. The film was shot rather than AI-generated, but generating, selecting, organizing, and editing alone also calls for a time margin.
For the empty-chair film, a five-shot plan might contain a view that establishes the chair, a view showing the person and cup, a closer view that makes the hesitation visible, a view of the decision, and a final image that lets its effect settle. Each shot has a job. If the crucial hesitation cannot be generated convincingly, a different angle—or a still image held for a carefully chosen duration—may communicate it better than repeated attempts at difficult motion.
Before production, distinguish three counts:
- Planned shots: the images the story needs in the edit.
- Candidate outputs: versions you expect to inspect and reject or keep.
- Available attempts: what your verified free access actually permits.
Do not set the third count by wishful thinking. If you can make still images but have no usable free video allowance, plan a short assembled from AI-generated images, cuts, and sound rather than assuming moving shots will become available. Be accurate about the result: editing motion into stills is not the same as generating moving footage, but it can still make an intentional film.
Treat the phone as the entire workspace
A phone-first production has to work on the device you possess—not just in a service’s promotional example. Corinne Podger’s Filming with a Smartphone is written for mobile video reporting. Its preparation checklist is useful here because storage, power, format, and an early test matter whether your source footage is recorded or generated.
Filming with a Smartphone - Mobile Journalism Manual
Read “Filming with a Smartphone” by Corinne Podger in the Mobile Journalism Manual for its audience-first planning and phone-readiness checks. Translate the filming advice into a generated-media workflow where appropriate.
In “Preparing for Your Video Shoot,” read the planning questions, from the story idea through the questions about format, essential shots, and sound. Then, under “Why a Storyboard Is Important for Filming,” read the shot-list advice. In “Mobile Video Journalism Checklist,” look under “Before You Leave for the Shoot” and read the first checks on familiarity, storage, and charging. For an AI film, the equivalent of checking footage before leaving a location is testing that you can save an asset, place it in your editor, and export a short sequence.
Make those checks concrete before you produce the full set of shots. Choose one delivery shape—vertical or horizontal—so you are not solving two compositions at once. On your phone, test one generated or imported image, one short clip if you intend to use video, a simple edit, and an export you can play back. Check that the text or action remains legible on the phone screen and that source files, editing files, and the export fit in available storage. Charge the device and account for the connection needed to generate or download assets.
A failed test is useful information. If the editor stalls with several clips, reduce clip length or use more stills. If an export loses detail, revise the delivery target to one the phone can produce reliably. A completed, watchable file at a modest resolution is a better first constraint than an unfinished plan for a higher-resolution film.
Budget cash and production time
For this lesson, a strict zero-cash rule means no subscription, purchased credits, paid trial that will renew, commissioned asset, or required accessory. It does not mean that every service with a free sign-up is suitable. Free allowances may run out; exports may be restricted or watermarked; rights may differ between personal and public use. Verify current terms before depending on a service, especially if you intend to release the film. We will examine those checks in detail later.
Time deserves an explicit ceiling too. Here is a planning allowance, not a promised completion time, for the five-shot example:
| Work | Allowance |
|---|---|
| Set the story, format, and shot ceiling | 1 hour |
| Test the phone-to-export workflow | 1 hour |
| Develop and select the core images | 2 hours |
| Attempt moving shots, if available; select usable material | 2 hours |
| Assemble picture and simple sound | 2 hours |
| Export and watch the complete film on the phone | 1 hour |
| Allow for failures and revision | 2 hours |
| Total reserved | 11 hours |
If moving shots are unavailable, the two-hour allowance can go toward improving still-image selection and editing instead. If you have only four hours in total, shorten the film and reduce the shot count before production. Do not remove the workflow test, the complete rough edit, or the final watch-through: those are what establish whether you have a film rather than a folder of assets.
The limits work together. A deadline may force fewer shots. A small free allowance may favor a story told with mostly stills. Limited storage may favor one format and fewer retained candidates. None of these decisions automatically makes the story less meaningful; they make its demands explicit.
A workable first-production boundary
For the empty-chair example, the production boundary could read:
Deliverable: one 30–45-second film, in one chosen screen format, playable as a single file on the phone. Story ceiling: one room, one visible character, two essential props, no required spoken dialogue, and no more than five planned shots. Cash ceiling: no purchases or subscriptions. Time ceiling: 11 reserved hours, including a two-hour margin. Fallback: if moving shots or a free sound option are not usable, tell the story with selected AI-generated stills and a deliberately simple edit.
This boundary is specific enough to make decisions. A striking generated street scene would not enter the film merely because it looks good; it creates a new setting without helping the chair-and-cup story. If the final gesture repeatedly fails, you can change how it is shown while keeping the same narrative decision. If a tool requires payment to remove an unacceptable export restriction, change the tool or the production approach rather than silently breaking the cash ceiling.
Scope is successful when it protects the story’s essential moment and leaves a credible path to a finished export. It is not a contest to use the fewest tools or shots.
Conclusion
A realistic zero-budget mobile plan sets a ceiling on what must appear, checks what the phone can actually produce, reserves time for failed attempts and editing, and keeps cash expenditure at zero without assuming free services are unlimited. Start with a complete, small story and test the route to a finished file early.
Next, you’ll turn these practical boundaries into a concise creative brief: who the first micro-short is for, what feeling and theme it aims to convey, and how you will recognize whether it works.