Microsoft-backed OpenAI is working on software that can generate minute-long videos based on text prompts, the company said Thursday.
The software, called Sora, is currently available for the red team to help identify flaws in the AI system, as well as for use by visual artists, designers and filmmakers to get feedback on the model, the company said in a statement.
“Sora is capable of generating complex scenes with multiple characters, specific types of movement, and precise subject and background details,” the statement said, adding that it can create multiple shots within a single video.
Challenge: The camera is directly facing the colorful buildings in Burano, Italy. An adorable dalmatian is looking out the window at the building on the ground floor. Many people walk and cycle along the canal streets in front of the buildings.
In addition to generating videos from text prompts, Sora can animate a still image, the company said in a blog post.
The video generation software follows OpenAI’s ChatGPT chatbot, which was released in late 2022 and created a buzz around GenAI with its ability to compose emails and write code and poems.
Last year, social media giant Meta Platforms boosted its Emu image generation model with two AI-based features that can edit and generate videos from text prompts.
The Facebook parent is also trying to compete with Microsoft, Alphabet’s Google and Amazon in the fast-moving generative AI universe.
Sora is in development, and the company adds that the model may confuse the spatial details of the challenge and have difficulty following a specific camera trajectory.
OpenAI said it is also developing tools that can tell if a video was generated by Sora.
