Recently I watched the oldest surviving movie, roundhay garden view, Which is from 1888. Four figures, two men and two women, walk around a yard with quick, jerky steps. This lasts for about two seconds.
I also recently watched some clips made in 2016 by researchers at the Massachusetts Institute of Technology and the University of Maryland, among them The first completely artificial-intelligence-generated videoEach is about a second long, In one, a blurry figure stands on a golf green, bent over at the waist, No one will confuse these videos or roundhay garden view For the effortless realism of contemporary cinema. And just as skeptics often deride AI video as useless, 19th-century critics described early cinema as “foolish curiosity,
Yet a recent agreement between Disney and OpenAI offers a glimpse of a different future. Starting in early 2026, the tech company’s video generator Sora will be able to create videos featuring more than 200 characters from Disney, Marvel, Pixar, and more. star wars Franchise. And Disney+ will stream a selection of user-generated clips.
On supporting science journalism
If you enjoyed this article, consider supporting our award-winning journalism Subscribing By purchasing a subscription, you are helping ensure a future of impactful stories about the discoveries and ideas shaping our world today.
Disney will also invest $1 billion in OpenAI and use its tools to create “new experiences for Disney+ customers,” according to one Joint press release from Disney and OpenAIAnnouncing the partnership, Disney CEO Robert Iger said the company would “thoughtfully and responsibly expand the reach of our storytelling through generative AI,” He also said in a recent earnings conference call that he intends to create content for subscribers Inside Disney+If you want to see Elsa and Cinderella kill Maleficent, you’ll be able to ask about that scene – although it may only last 20 seconds,
If this is the beginning of AI TV on demand, I wonder how long it will take for these clips to reach 20 minutes or an hour, given the environmental burden and computing costs. Many people believe that it is impossible, but I think there will be very few people who will see roundhay garden scene Preview the great train robbery, A 12-minute milestone of silent cinematography from 1903, very short gone With the Wind—or streaming.
The challenge of image generation lies in how today’s systems work. They are built on diffusion, a technique that starts with “noise” that is gradually refined into an image. Draw an image of a person standing in the fog. The AI ​​essentially removes the blur and inserts new pixels in repeated passes until a coherent shape appears. The cost of each pass increases to refine the generated image.
The video is even more challenging. The series of images must be coordinated so that facial features do not change and coffee mugs do not disappear. Millions of pixels are changing in one second of high-definition video. during a lack of privacy “We discovered how painful it is to work with video data. There are a lot of pixels in these videos,” said Bill Peebles, an OpenAI researcher who helped develop Sora, at a hackathon hosted by AI community hub AGI House.
To manage pixels, OpenAI’s system Compresses the video into a simplified version that keeps important information. It then treats it like bread – cutting it into pieces and then dividing it into cubes. This allows the model to coordinate all the cubes with each other, in much the same way as the models powering ChatGPT combine all the words in the response.
The jump from seconds to minutes is so painful because the more frames you add, the more information the model has to take into account. As videos become longer, discrepancies increase. True “on-demand” AI TV would also require cuts between scenes. If every Disney+ user was requesting it with near-term technology, the cost would be staggering.
Researchers are looking for more efficient methods. One model is to divide the task into steps. “Instead of rendering or preparing the entire video at once, you prepare frame by frame,” says Tianwei Yin, a research scientist at AI image editing start-up Rave. CausVid video-generation software“At each step, your calculations are limited to a much smaller part rather than the whole thing, and this enables you to go longer,”
Yin believes that systems will reach five-minute generation more efficiently by next year and, through the integration of various existing AI technologies, they could reach one hour shortly after. Others have echoed this optimism. recently bbc interviewGoogle CEO Sundar Pichai described the possibility of high school students creating feature-length AI films in the coming years. Cristóbal Valenzuela, CEO of AI-video-generation company Runway, told el pais Earlier this month, “It’s still not possible to have 60 or 90 minutes with consistent characters and story. But it will happen soon.” He further said that seeing AI videos generated in real-time is also on the horizon.
The path from curated fan clips to feature-length films will go through some unsurprising innovations, not to mention negotiations over how to pay the creatives whose work it promotes. And although the financial burden of AI video seems prohibitive, millions of people globally are involved in the production and training of AI models, and the costs of the technologies are generally low. For example, in 1998 bandwidth was extremely expensive—it cost About $1,200 per megabit per second (Mbps) monthly for large networks—but reported costs were the lowest until 2025 $0.05 per Mbps monthlyReduction of 99.996 percent. This change made streaming on Disney+ or Netflix possible.
The cultural trajectory of new mediums is far more difficult to imagine, and resistance is often intense. The poet Charles Baudelaire criticized photography in 1859 for its lazy realism, which distanced art from the imagination. In previous centuries, “both skeptics and partisans compared photography to painting and moving pictures to theater.” Present-day scholar Ruben de Latour wroteIt appears that we are in an even more complex moment, What seems certain is that, as in the past, technology will evolve rapidly, allowing millions of creators to test possibilities that we cannot yet predict,
It’s time to stand up for science
If you enjoyed this article, I would like to ask for your support. scientific American He has served as an advocate for science and industry for 180 years, and right now may be the most important moment in that two-century history.
i have been one scientific American I’ve been a member since I was 12, and it’s helped shape the way I see the world. Science Always educates and delights me, and inspires a sense of awe for our vast, beautiful universe. I hope it does the same for you.
if you agree scientific AmericanYou help ensure that our coverage focuses on meaningful research and discovery; We have the resources to report on decisions that put laboratories across America at risk; And that we support both emerging and working scientists at a time when the value of science is too often recognised.
In return, you get the news you need, Captivating podcasts, great infographics, Don’t miss the newsletter, be sure to watch the video, Challenging games, and the best writing and reporting from the world of science. you can even Gift a membership to someone.
There has never been a more important time for us to stand up and show why science matters. I hope you will support us in that mission.
