OpenAI's 'Sora' text-to-video AI set for public release later this year

OpenAI is gearing up to make its groundbreaking text-to-video AI model, 'Sora,' publicly available later this year, marking a significant milestone in artificial intelligence innovation.

Originally introduced earlier this year, "Sora" captivated viewers with its ability to turn text prompts into immersive 60-second videos that are as realistic as the actual footage.

In an interview with The Wall Street Journal, OpenAI's Mira Murati confirmed the imminent release of 'Sora,' stating it will be available "definitely this year" and potentially within "a few months." Moreover, users can anticipate the ability to "edit" output by requesting fixes directly, although audio functionality will not be supported initially.

Murati shed light on the model's training data, revealing that it consists of publicly available video content and licensed material from Shutterstock. However, specific details regarding the inclusion of YouTube videos remain undisclosed.