OpenAI introduces AI model that turns text into video – February 15, 2024 at 11:44 pm

Microsoft-backed OpenAI said Thursday it is working on software that can create one-minute videos based on text prompts.

The software, called Sora, is currently available to the red team, helping to identify errors 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 able to generate complex scenes with multiple characters, specific types of movements, and precise subject and background details,” the statement said, adding that it can capture multiple shots in a single video.

In addition to creating videos from text prompts, Sora can also animate a still image, the company said in a blog post.

The video creation software follows OpenAI's ChatGPT, which was released in late 2022 and generated buzz around GenAI with its ability to compose emails and write codes and poems.

Social media giant Meta Platforms last year expanded its image generation model Emu with two AI-based features that can edit and create videos based on text prompts.

Facebook's parent company also wants to compete with Microsoft, Alphabet's Google and Amazon in the rapidly transforming world of generative artificial intelligence.

Sora is a work in progress, with the company adding that the model can get confused by the spatial details of the router and have difficulty following a specific camera path.

OpenAI said it is also developing tools that can differentiate whether a video was created by Sora.