News

ChatGPT’s creator OpenAI has announced a new AI model called Sora that can create “realistic and imaginative scenes from text instructions” that are up to one minute long. According to OpenAI,   is able to create complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background to generate compelling characters that express vibrant emotions. “The model understands not only what the user has asked for in the prompt, but also how those things exist in the physical world,”  . , Sora can sample widescreen 1,920 x 1,080 videos, vertical 1,080 x 1,920 videos and everything inbetween. This lets Sora create content for different devices directly at their native aspect ratios. OpenAI does admit that there are weaknesses with the current model, such as confusing spatial details like left and right or cause and effect. For example, a person might take a bite out of a cookie, but afterward, the cookie may not have a bite mark. Sora is currently undergoing safety testing with red teamers — domain experts in areas like misinformation, hateful content, and bias. At the same time, OpenAI will leverage existing safety methods that were built for products that use DALL·E 3, which are applicable to Sora as well. So Sora will reject text input prompts that are in violation of our usage policies, like those that request extreme violence, sexual content, hateful imagery, celebrity likeness, or the IP of others. Sora isn’t the first generative AI model for images, but the major difference between this and   from Microsoft, is that the latter creates static images while Sora can create moving images.