OpenAI Launched a new AI tool on Thursday that can create videos from textual descriptions.The new model, called Sora after the Japanese word for “sky”, can generate realistic footage up to a minute long that matches the user’s specifications on both content and style. The model can also synthesize a video from a single image or augment existing footage with new elements, according to a company blogpost.
The company’s blogpost states: “We’re teaching AI to understand and simulate the physical world in motion, with the goal of training models that help people solve problems that require real-world interaction.”
One of the videos showcased by the company was based on the following prompt: “A movie trailer featuring the adventures of the 30-year-old space man wearing a red wool knitted motorcycle helmet, blue sky, salt desert, cinematic style, shot on 35mm film, vivid colors.”
The company announced that it had granted access to Sora to a select group of researchers and video creators. The experts would “red team” the product – test it for potential misuse or violation of OpenAI’s terms of service, which forbid “extreme violence, sexual content, hateful imagery, celebrity likeness, or the IP of others”, as per the company’s blogpost.
The company is limiting access to researchers, visual artists and film-makers only, though CEO Sam Altman shared some video clips he said were made by Sora on Twitter after the announcement. The videos have a watermark to indicate they were made by AI.
