OpenAI's new text-to-video model, Sora, is now under review and in the testing phase. Sora can generate videos from textual prompts and according to its developers, it can also generate videos from images.