The AI company behind ChatGPT announced the new tool, Sora, on Thursday. It's designed to create realistic videos based on user prompts. People can offer up existing images with their prompts or longer videos, the company said.
The product is still getting safety tested among select users, and isn't available yet to the general public. But Altman gave a few people a chance to see their prompts brought to life using Sora.
"We'd like to show you what sora can do, please reply with captions for videos you'd like to see and we'll start making some," Altman tweeted shortly after the company announced the new product on Thursday.
One X user asked Altman to generate a video of "Two golden retrievers podcasting on top of a mountain."
Altman shared the results from Sora and the video was accurate! It was also detailed: The dogs had headphones, mics, and even a red picnic blanket to do their podcasting from.
Another X user requested to see "different animals as athletes riding bicycles." In response, Sora delivered a bunch of sea creatures, some with human-like legs, biking on water.
But, as one user commented, "the turtle can't reach the pedals ."
Another person asked to see a video in which a "half duck half dragon flies through a beautiful sunset with a hamster dressed in adventure gear on its back."
Sora again delivered, even if the hamster's legs were rather long and the duck-dragon appeared to be flying backwards.
- OpenAI's Sam Altman unveiled a new video-generating tool called Sora.
- Sora creates realistic videos based on user prompts, including images and longer videos.
- The tool, currently undergoing safety tests, generated a user-requested video of podcasting golden retrievers