IBL News | New York
OpenAI announced yesterday the launch of its video model Sora, an AI tool for creating realistic clips from text. It lets users generate videos up to 20 seconds long. It includes dropping images in as prompts and a timeline editor that allows users to add new prompts at specific moments in a video.
“We don’t want the world just to be text,” OpenAI CEO Sam Altman said in a live-streamed announcement Monday. “Video is important to our culture,” Altman added.
The San Francisco-based research lab, backed by Microsoft, said that this current version of Sora, named Sora Turbo, is faster than the model showed in February.
Sora has been released as a standalone product at Sora.com to ChatGPT Plus and Pro users.
ChatGPT Plus offers up to 50 videos at up to 480p resolution, up to 20 seconds long, or fewer videos at 720p monthly. These videos are in widescreen, vertical, or square aspect ratios at no additional cost. Users bring their assets to extend, remix, and blend or generate entirely new content from text.
Meanwhile, ChatGPT Pro offers up to 500 monthly videos at up to 1080p resolution and longer duration.
OpenAI developed new interfaces to make it easier to prompt Sora with text, images, and videos. Its storyboard tool lets users precisely specify inputs for each frame.
OpenAI acknowledged that “the version of Sora we are deploying has many limitations, and it often generates unrealistic physics and struggles with complex actions over long durations.” “Although Sora Turbo is much faster than the February preview, we’re still working to make the technology affordable for everyone.”
The Sora model is blocking particularly damaging forms of abuse, such as child sexual abuse materials and sexual deepfakes.
Sora could also violate many creators’ rights, experts said.
• Sora features
• Video
