Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallSome links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
On March 25, 2024, OpenAI published “Sora: first impressions,” a showcase of short works by seven invited creators using its then-unreleased video-generation model. The examples offered an early look at Sora as a tool for visual experimentation and concept development—not a public launch, an independent performance test, or proof that the model could make finished films on its own.
An invited showcase, not a public launch
OpenAI had introduced Sora in February 2024 as a research-preview model. It said it was giving selected visual artists, designers and filmmakers access to gather feedback and explore how the system might fit into creative workflows. The March 25 post brought together work and comments from some of those outside creators.
At the time, ordinary users could not sign up and use Sora. Invited access for selected creators and safety testers was different from general public availability, and neither amounted to a commercial product launch. OpenAI called the videos a few examples and said the model still needed improvement. The announcement should be read in that limited context.
OpenAI described Sora as able to generate video from text, animate still images, extend existing footage and fill in missing frames. Its February announcement also discussed generating different durations, resolutions and aspect ratios. Those were the model’s stated capabilities; the March reel did not independently establish how reliably they worked across everyday prompts.
#1 Best Overall
Who made the videos—and what they explored
The showcase featured seven creators or creative groups. Their examples are most revealing when considered as different kinds of work rather than as a single demonstration that Sora could replace a production pipeline.
shy kids: “Air Head”
Toronto multimedia company shy kids—Walter Woodman, Sidney Leeder and Patrick Cederberg—made Air Head, a short film built around a character with a balloon for a head. Its surreal premise illustrates how a generator might help visualize something difficult or costly to stage physically. But OpenAI said the artists edited the videos and could modify Sora-generated material. The film was an artist-directed work incorporating Sora, not evidence of an untouched, one-prompt result.
Paul Trillo: surreal visual exploration
Multidisciplinary artist, writer and director Paul Trillo presented experimental imagery. His comments emphasized the freedom to explore ideas without the usual constraints of time, budget, locations or coordinating collaborators. That makes his work a useful example of ideation and previsualization—especially for impossible or dreamlike images—not a demonstration of conventional, long-form narrative production.
Nik Kleverov and Native Foreign: concepts for brands
Creative director Nik Kleverov, co-founder of the Emmy-nominated agency Native Foreign, described using Sora to visualize ideas and iterate quickly for brand partners. Of the examples, this most clearly points to a commercial-development use: an agency can show a client a possible visual direction before committing to a shoot or a more complete production.
That is different from delivering a finished advertisement. A client-facing concept still needs direction and review; final work may also require editing, sound, compositing, rights checks, brand approval and conventional production. The announcement did not establish that Sora eliminated those stages or their costs.
August Kamp: an artist’s exploratory tool
Musician, researcher, creative activist and multidisciplinary artist August Kamp approached Sora as part of an artist-led process. The example suggests a role like a visual sketchbook: exploring images and possibilities around creative work, rather than treating generation as an automatic substitute for a camera crew or other collaborators.
Josephine Miller and Oraar Studio: digital design possibilities
Josephine Miller, co-founder and creative director of Oraar Studio, works across 3D visuals, augmented reality and digital fashion. Her inclusion highlights possible experiments with garments, materials, movement and designed environments—visual questions that might be explored before building a detailed 3D or physical-production workflow. The showcase did not establish how consistently Sora could preserve exact design details from one shot to another.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Don Allen Stevenson III: worlds for AR and XR
Digital AR/XR artist, speaker and consultant Don Allen Stevenson III brought a perspective grounded in digitally constructed worlds. For immersive-media creators, generated clips could help test an atmosphere, environment or speculative visual idea. A clip that suggests a world, however, is not the same as a coherent interactive environment or a production-ready set of assets.
Alex Reben: from video toward 3D forms
Sculptor and artist-in-residence at OpenAI Alex Reben was interested in extending AI-generated video toward 3D forms and models. His example points to a broader creative question: whether moving images can contribute to workflows for imagining or building objects and environments, rather than ending as video alone. The post was an exploration, not proof that Sora itself was a dependable 3D-asset tool.
OpenAI’s descriptions and the creators’ comments appeared in the company’s announcement. They are useful first impressions from selected participants, but they are not independent reviews or a representative sample of all Sora outputs.
The human work behind “made with Sora”
OpenAI explicitly said the showcased videos were edited by the artists, who had freedom to modify Sora-generated material. That disclosure matters: the final works involved human choices about what to generate, what to keep, how to shape it and how to present it. “Sora-assisted” or “created with Sora” is more accurate than saying the model independently made each film.
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 minuteGenerative video may shift work rather than remove it. Instead of planning every shot around a physical location, crew and camera setup, creators may spend more time prompting, rerunning generations, selecting usable footage, maintaining visual continuity, editing and cleaning up results. Lowering the barrier to a first visual draft is not the same as removing the judgment and craft needed to turn it into a finished piece.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What the reel did—and did not—demonstrate
The strongest promise in the selected examples was the ability to propose images that might be awkward, expensive or impossible to capture conventionally. For filmmakers, that could mean mood pieces or previsualization; for agencies, rapid concept exploration; for visual artists, an expanded space for experimentation; and for digital designers, early studies of constructed worlds and forms.
But a curated reel cannot establish typical performance. The clips were selected for a company announcement, not collected under an independent benchmark protocol. A visually compelling short sequence does not prove reliable character continuity, precise blocking, realistic physics, complex action over long durations, or the ability to complete a coherent film. OpenAI itself said Sora still needed substantial improvement.
That distinction is especially important because surreal imagery can make a compelling demonstration while sidestepping the demands of realistic dialogue, sustained action or exact continuity. The announcement did not show that every prompt would yield comparable footage, that the outputs were repeatable, or that production savings outweighed the time spent on selection and post-production.
It also did not resolve questions about training data, consent, attribution, ownership or compensation. Contemporary reporting by VentureBeat noted that OpenAI had not provided a complete account of the specific training data. The creator examples can illustrate a possible collaboration between artists and a generative model, but they do not settle wider concerns about how such systems are trained or how their effects on creative labor should be handled.
How the showcase fits Sora’s later history
- February 2024: OpenAI introduced Sora as a research-preview model and described its early capabilities and selected-creator access.
- March 25, 2024: OpenAI published the first-impressions showcase featuring outside creators’ work.
- December 9, 2024: OpenAI announced Sora Turbo as a standalone product for eligible ChatGPT users, a later step distinct from the invited showcase.
- September 30, 2025: OpenAI published a system card for Sora 2.
- April 26, 2026: OpenAI says the Sora web and app experiences were discontinued.
- September 24, 2026: OpenAI’s help documentation lists this as the scheduled discontinuation date for the Sora API.
The current status is therefore different from the access situation in 2024. The original Sora web and app experiences are no longer available as of April 26, 2026, and the API is scheduled to end on September 24, 2026, according to OpenAI’s discontinuation notice. The showcase remains useful as a historical snapshot, not as an invitation to sign up for the original product.
Why the announcement mattered
The March 2024 post made the creative promise of video generation tangible by showing work from people outside OpenAI. It suggested that artists and production teams might use a model to make unusual images, test directions and communicate visual ideas earlier in a project. It also exposed the limits of what a promotional reel can prove: the works were selected, edited and presented by their creators, access was restricted, and the broader questions of reliability, rights and labor remained open.
For documentary and film audiences, the most useful takeaway is not that a machine had learned to make films autonomously. It is that a new tool might change how visual ideas are proposed and developed—and that the human decisions behind the finished work still matter.
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.
