OpenAI Unveils Sora: The AI That Creates Videos from Text

illustration by Andy McNally

Sora is a new AI model from OpenAI that can create realistic and imaginative videos from text prompts. OpenAI says that Sora can generate videos up to a minute long at 1080p resolution while maintaining visual quality and adherence to the user’s prompt.

sketchnote by Andy McNally

Text-to-Video Generation
Sora can generate complex scenes that can include multiple characters, specific types of motion, and accurate details of the subject and background from simple text prompts.

Bate-Papo, an AI chatbot, eating popcorn illustration by Andy McNally

Understanding the Physical World
Sora understands language and knows how things should look and act in the real world. It can make videos with characters that seem real and show emotions. Sora can also keep the same visual style and characters throughout the video.

Video from Still Images
Sora can use an existing still image and animate its contents, as well as extend existing videos or fill in missing frames with accuracy and attention to small detail.