OpenAI’s new Sora model can produce one-minute videos from text instructions

This innovation artificial intelligence It is considered an indicator of progress in technology. The Sora model can take text-based instructions and transform them into dynamic and visually rich videos. This allows users to easily visualize their creative ideas and produce impressive content. This development, which is considered as a great innovation in the world of technology, has a significant impact on both the entertainment industry and the education ve
marketing It opens new doors in many areas such as.

Prompt: “Several giant wooly mammoths
approach treading through a snowy meadow, their long wooly fur
lightly blows in the wind as they walk, snow covered trees and
dramatic snow capped mountains in the distance, mid afternoon light
with wispy clouds and a sun high in the distance… pic.twitter.com/Um5CWI18nS

— OpenAI (@OpenAI)
February 15, 2024

OpenAI The newly developed Sora model can produce one-minute high-quality videos based on text instructions. in japanese Sora, which means “sky”, will not be available for general use for now. Instead, OpenAI presents this model to a small group of academics and researchers who will evaluate potential abuse and harms.

Sora is able to produce complex scenes involving multiple characters, specific types of movements, and accurate details of the plot and background. OpenAI One example video he shared shows a couple walking through a snowy Tokyo city, with cherry blossom petals and snowflakes floating around them. Another video shows realistic-looking woolly mammoths walking through a snowy meadow against the backdrop of snow-capped mountain ranges. The model exhibits this functionality thanks to its ability to accurately interpret text instructions. However, just as we see other AI image and like video producers, Sora isn’t perfect. For example, an instruction requesting a video of a Dalmatian looking out the window and people “walking and cycling along the canal streets” completely ignores the people and streets in the video.

Introducing Sora, our text-to-video
model.

Sora can create videos of up to 60 seconds featuring highly
detailed scenes, complex camera motion, and multiple characters
with vibrant emotions. https://t.co/7j2JN27M3W

Prompt: “Beautiful, snowy… pic.twitter.com/ruTEWn87vf

— OpenAI (@OpenAI)
February 15, 2024

Sora is not the first model to create videos from text. Meta, Google and Runway Other companies such as have also worked on text-based video tools or made them available to the public. However, there is currently no other tool that can produce video up to 60 seconds. In addition, unlike other models, Sora creates videos entirely at once, instead of combining them frame by frame. This ensures that the subjects of the video remain the same throughout the video, even if they are temporarily out of view.

Text-based video tools Its rise has raised concerns that realistic-looking fake images can be more easily created. is a professor of artificial intelligence at the University of Washington who works to identify misinformation in political campaigns.
True Media’s “I’m absolutely terrified that something like this could narrowly affect an election,” founder Oren Etzioni said. Creative AI in general is receiving pushback from artists and creative professionals about using technology to change jobs.

OpenAI, He said that he tested the tool by working with experts in areas such as misinformation, hate speech and bias before making it available to the public. The company is developing tools that can detect videos produced by Sora and adding metadata to the videos produced for easier detection. Company on how Sora was trained The
New York Times’a, It declined to provide information other than stating that it used “public domain videos” and videos licensed from copyright holders.