What’s actually slowing this PC down?
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPick the symptom - the matching free tool is one click away.
ThatPainter is reader-supported. When you buy through links on our site, we may earn an affiliate commission. Learn More
AIFF 2025 did not prove that artificial intelligence can replace filmmakers. The Runway-hosted festival in Santa Monica, California, offered a more complicated demonstration: generative AI can help artists visualize impossible shots, explore images rapidly, and make certain kinds of filmmaking accessible to small teams. But the strongest work still depended on human perspective, judgment, editing, and purpose—and the festival could not resolve the harder questions about consent, copyright, labor, credit, or control.
The event, reported by Ars Technica on June 24, 2025, screened 10 short films made entirely or partly with generative-AI tools. Their differences matter more than the label “AI film.” Some treated the technology as spectacle; others used it to express memory or solve a practical production problem.
A festival built around a disputed question
AIFF 2025 brought together AI enthusiasts, professional creatives, and film audiences for a public showcase of short films. Runway hosted the event in Santa Monica, and its eligibility was broad: a film could be made entirely with Runway or another AI system, or use AI as one important part of a conventional workflow.
That breadth makes “AI-made” an unstable description. It can mean generated footage, AI-assisted visual effects, synthetic voices, rotoscoping, previsualization, or a conventional film containing a few generated shots. Those are different creative and labor arrangements, with different legal and ethical consequences.
The judging panel included filmmakers, studio and VFX executives, and technology-industry representatives. Runway CEO Cristóbal Valenzuela described the company’s approach as integrating AI into studio workflows, including work with creative organizations and relationships mentioned in the coverage with Lionsgate and AMC Networks. Those are company positions and industry signals—not proof that every studio workflow is cheaper, safer, or better because AI is involved.
Nor should the festival be treated as a complete official catalogue. The available reporting describes 10 films observed by the reporter, but does not establish a complete record of submission numbers, budgets, audience metrics, or every eligibility and disclosure rule.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11What the films revealed
Fragments of Nowhere: when the tool becomes the subject
Vallée Duhamel’s Fragments of Nowhere used transformations, distorted bodies, and impossible physical movement. It demonstrated one of generative video’s obvious attractions: the ability to produce imagery that would be expensive, technically difficult, or impractical to capture conventionally.
But visual impossibility is not automatically cinematic meaning. Morphing bodies, dreamlike transitions, and anatomical instability can become a recognizable “AI aesthetic”—a mixture of dreamcore, horror, and surreal spectacle. The important question is whether the transformation serves a character, idea, or emotional progression, rather than merely displaying what the model can do.
More Tears than Harm: specificity is stronger than novelty
Herinarivo Rakotomanana’s More Tears than Harm was described as a rotoscope-style sensory collage rooted in childhood memories of growing up in Madagascar. Its force came less from technical strangeness than from a personal and culturally specific point of view.
This distinction is central to judging AI-assisted art. A system can generate an unfamiliar image, but unfamiliarity is not the same as lived experience. Memory, place, cultural context, and emotional intention give images a reason to exist. They also give the artist something to be accountable for.
Jailbird: AI as a practical production solution
Andrew Salter’s Jailbird, the festival’s runner-up, used AI to depict a chicken’s point of view in a documentary-style story about chickens placed in UK prisons as companion animals. Here, the technology had a clear production rationale: obtaining certain shots would have been difficult or impossible for a small-budget production.
That is a more useful model for understanding AI’s near-term role. The question is not whether a machine can independently make a film. It is whether a filmmaker can use a generative system to achieve a specific shot, perspective, or transition that would otherwise require resources the production does not have.
Total Pixel Space: the philosophical centerpiece
Jacob Adler’s Total Pixel Space won the Grand Prix. Its central idea is that every possible image exists within a vast mathematical space of pixel arrangements, and that artists—including AI-assisted artists—select or discover possibilities rather than create from absolute nothing.
The argument gives curation a serious role in creativity. A filmmaker chooses a subject, develops a treatment, requests or creates images, rejects most attempts, sequences shots, controls pacing, and gives the finished work an interpretation. Generative AI expands the range of possibilities available during that process.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →But this is a philosophical claim, not a settled definition of artistic authorship. The fact that images can be represented mathematically does not make all images artistically equivalent. Human expression may also reside in lived experience, trained judgment, physical effort, intention, cultural knowledge, and personal risk. Nor does the “everything exists in pixel space” argument answer who supplied the training material or who should be paid when a system draws on creative work without permission.
The film’s win also contained an unavoidable institutional tension: an AI company’s festival awarded a film that makes the case for AI art. That does not invalidate the film, but it is relevant context when evaluating the claims made around the event.
“AI film” describes a spectrum, not a single process
For filmmakers and audiences, the most useful classification is to identify what the system actually did:
- Pre-production visualization: concept development, storyboards, mood boards, and previsualization before expensive production decisions.
- Editorial and post-production assistance: shot extension, cleanup, organization, rotoscoping, compositing, and other modifications to existing material.
- Generative creation: producing new images, shots, voices, performances, or environments.
- Synthetic replacement: substituting for actors, designers, editors, VFX artists, illustrators, or other specialists.
- Human-led assembly: selecting outputs, directing visual choices, editing, designing sound, writing, and shaping the final narrative.
These categories should not be collapsed. Using AI to extend a shot by two seconds is not the same as generating an actor’s performance. Using a model for storyboards is not the same as replacing a storyboard artist. A film’s credits and disclosure should make those distinctions visible.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteWhere AI is already useful
The festival and the surrounding industry discussion point to constrained uses where generative systems can reduce friction:
- rapid visual exploration during development;
- previsualizing scenes and camera positions;
- extending or altering shots;
- certain visual-effects and rotoscoping tasks;
- creating otherwise expensive points of view;
- testing concepts before committing crew, locations, sets, or equipment;
- helping small or first-time teams attempt ambitious visual ideas.
These benefits are real but should not be described as automatic cost savings. Generation may be inexpensive while the complete workflow is not. Iteration, failed outputs, continuity repair, compositing, editing, sound, rights review, delivery, and human supervision can consume substantial time. A lower model bill does not necessarily mean a lower production cost.
AI can also lower some barriers without making filmmaking equally accessible. Users still need software, computing access, visual judgment, writing ability, production knowledge, distribution, and often paid subscriptions. The technology may broaden participation while simultaneously increasing competition and putting downward pressure on creative rates.
Why Hollywood is divided
Supporters frame AI as another major production technology. On this view, film history is full of tools that changed what crews could do, and generative systems may give small teams capabilities once reserved for large productions. The most immediate value may be faster visualization and selective assistance, not autonomous storytelling.
Critics see a different risk. If models were trained on copyrighted creative work without permission, the tool may embody an uncredited extraction of labor. If generated imagery, voices, performances, or design work replaces specialists, efficiency can weaken employment and bargaining power. A workflow marketed as assistance can become substitution when employers decide that fewer people are needed.
The disagreement is therefore not simply artists versus technology. Some working creatives already use these tools and want them improved. Others see the same systems as a threat to the economic foundation of their professions. The distribution of gains matters as much as the existence of the tool: who owns the model, sets access and pricing, receives the productivity benefit, and carries the risk?
Copyright, consent, and responsibility
Two copyright questions are often incorrectly treated as one.
1. What trained the model?
A project may need to ask what material was used to train a system, whether it was licensed, public-domain, scraped, or obtained under disputed terms, whether creators were informed or compensated, and whether they can opt out or demand removal. Claims about training data must be distinguished from final legal determinations.
2. What does the output do?
A generated result may raise separate questions if it reproduces protected expression, imitates a living artist, uses a recognizable character or franchise, or copies an actor’s likeness or voice. Responsibility may be divided among the model provider, studio, producer, and user, depending on the jurisdiction and contract.
Runway executives have emphasized output-side liability and offered indemnification to studio customers, according to the Ars Technica report. That should be understood as a company position and contractual allocation of risk, not a universal guarantee that every use is lawful. Indemnification may have conditions, exclusions, limits, and no effect on obligations owed to people whose work or likeness is used.
For a real production, teams should document source assets, permissions, prompts, model versions, generated material, human edits, approvals, and final credits. They should confirm commercial-use terms, upload privacy, model-training policies, retention, output warranties, and the treatment of voices, likenesses, and performances. Project-specific legal review remains necessary.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Can an AI-assisted film be human expression?
There is no useful binary answer. An AI-assisted film can contain substantial human expression through its subject, script, treatment, cultural context, prompt design, iteration, shot selection, direction, editing, sound, pacing, narrative structure, and final presentation.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Yet human involvement alone does not settle authorship. The decisive question is which expressive elements were determined by people and which were supplied by the system. A director who develops a personal story, controls the visual language, rejects weak generations, edits the sequence, and integrates sound has made different kinds of contributions from someone who accepts an unedited sequence after a single prompt.
Disclosure should therefore describe the process rather than use “AI-made” as a blanket label. Audiences and collaborators deserve to know whether AI generated footage, altered existing footage, supplied a voice, created a digital double, assisted editing, or appeared only in previsualization.
The most defensible conclusion is conditional: AI-assisted filmmaking can be human expression, but not every AI-generated sequence is equally expressive, original, attributable, or ethically defensible.
How to judge an AI film beyond technical novelty
- Narrative purpose: Does AI enable the story, or is it merely being displayed?
- Human specificity: Is there a distinctive perspective, memory, culture, or emotional point of view?
- Consistency: Do characters, props, lighting, geography, and body mechanics remain coherent?
- Editorial intelligence: Are the strongest moments selected and shaped thoughtfully?
- Technical appropriateness: Does AI solve a genuine production problem?
- Transparency: Are the tools and extent of their use disclosed?
- Labor impact: Does the workflow augment specialists or silently replace them?
- Rights provenance: Are training materials, likenesses, voices, and source assets responsibly handled?
- Audience value: Does the finished work create an experience that conventional methods could not achieve as effectively?
- Reproducibility: Is the result controllable and repeatable, or the product of a lucky generation?
This framework separates technical novelty from cinematic achievement. A film may be visually impressive but dramatically empty. Conversely, the best use of AI may be the one viewers barely notice because it serves the story rather than competing with it.
The future will be decided by institutions, not slogans
AIFF 2025 is most valuable as a case study in transition. It shows that generative tools are already entering filmmaking through previsualization, selected post-production tasks, impossible shots, and low-budget experimentation. It also shows why “AI is just a tool” is incomplete: tools exist inside systems of ownership, labor, contracts, training data, and distribution.
The lasting questions are practical. Will workers share in efficiency gains? Will performers control their likenesses and voices? Will artists receive consent and compensation when their work helps build commercial systems? Will audiences receive meaningful disclosure? Can productions preserve continuity, accountability, and cultural specificity? Can filmmakers leave a vendor if its pricing, model behavior, or license changes?
Whether generative AI becomes a valuable filmmaking instrument or a mechanism for uncredited substitution will depend less on declaring whether machines are creative than on the rules built around them. Contracts, court decisions, labor agreements, disclosure standards, and compensation structures will determine who gets to create—and who gets paid—inside the new workflow.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
