AI isn't uncommon nowadays; we already have ChatGPT and Google Bard for generative writing, and various AI software for photo generation—and now, we're going to have AI for generative videos as well.
In an exclusive interview with The Wall Street Journal, OpenAI's Chief Technology Officer, Mira Murati said that the text-to-video AI model would be available to public this year.
Sora was first teased by OpenAI in February. The AI model is capable of generating hyperrealistic scenes generated from simple text prompts, targeted at visual artists, designers, and filmmakers. Murati has also revealed plans to integrate audio to amplify the realism of Sora's creations.
As for where Sora's training data comes from, OpenAI has remained strangely tight-lipped, only disclosing that publicly available or licensed data like Shutterstock was used. Besides that, similar to DALL-E's policies, Sora won't produce images of public figures and will feature watermarks to distinguish its creations from reality.
Is this a milestone for creative content, or is the future of visual content threatened? Let us know your opinions!