OpenAI says its new text-to-video GenAI model, Sora, can generate photorealistic videos and extend videos users have made based on prompts

source