OpenAI’s text-to-video generator, Sora, is set to become publicly available later this year. Initially released to a select group of visual artists, designers, and filmmakers, Sora can create hyperrealistic scenes from text prompts. OpenAI CTO Mira Murati indicated that the release could happen within a few months and mentioned plans to eventually incorporate audio and allow users to edit the generated content. While specific training data details were not disclosed, Murati confirmed a partnership with Shutterstock. OpenAI aims to offer Sora at a cost comparable to its AI text-to-image model, DALL-E, despite Sora’s higher operational expenses.
Read more at: www.theverge.com