OpenAI is on the verge of releasing a new generative AI tool called Sora, according to an interview with the company’s CTO, Mira Murati, in The Wall Street Journal. Sora allows users to create realistic and imaginative videos simply by providing a text description.
The much-anticipated tool is expected to be available “this year” within a few months. While it won’t include audio initially, Murati highlights plans to incorporate audio for a more immersive experience in the future. Recognizing the potential for inaccuracies in AI-generated content, OpenAI also intends to provide editing functionalities for users to refine the videos.
Sora’s ability to handle complex scenes with multiple characters and specific motions. Murati emphasizes OpenAI’s goal of making Sora a flexible creative tool, showcasing its potential through examples like an animated bull in a china shop and a mermaid reviewing smartphones.
Although Sora is reportedly more expensive to operate, OpenAI aims to offer it at a similar cost to their popular text-to-image model, DALL-E, upon public release. This accessibility could open doors for a new wave of video creation fueled by user imagination and powered by AI.
