‘Bicycle Race On Ocean’ To ‘Futuristic Drone Race At Sunset’, OpenAI’s Latest 'Sora' Tool Can Do It All

OpenAI has revealed its latest generative text-to-video tool, Sora. As per OpenAI’s blog post, “Sora can generate videos up to a minute long while maintaining visual quality and adherence to the user’s prompt.” Currently, it is not available to the public as it is being tested by red teamers to “assess critical areas for harms or risks” in areas like “misinformation, hateful content, and bias.” Announcing the product online, Sam Altman, the CEO of OpenAI, asked netizens to give him their prompts and he would share the results generated by Sora.

On Thursday, Altman said, “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!” He added, “Don’t hold back on the detail or difficulty!”

Soon people came up with bizarre prompts like, “A bicycle race on the ocean with different animals as athletes riding the bicycles with drone camera view”, “A instructional cooking session for homemade gnocchi hosted by a grandmother social media influencer set in a rustic Tuscan country kitchen with cinematic lighting”, and “a wizard wearing a pointed hat and a blue robe with white stars casting a spell that shoots lightning from his hand and holding an old tome in his other hand.” Altman responded to many such prompts, with highly accurate results.

An X user gave a highly detailed prompt and wrote, “A street-level tour through a futuristic city which is in harmony with nature and also simultaneously cypherpunk/high-tech. The city should be clean, with advanced futuristic trams, beautiful fountains, giant holograms everywhere, and robots all over. Have the video be of a human tour guide from the future showing a group of extraterrestrial aliens the coolest and most glorious city that humans are capable of building.” Altman responded to this prompt with a 10-second clip.

Despite its impressive results, the OpenAI team admits that Sora still has many weaknesses. In its blog post, it wrote that Sora “may struggle with accurately simulating the physics of a complex scene, and may not understand specific instances of cause and effect.” It added that the model may confuse “spatial details of a prompt, for example, mixing up left and right.”

The OpenAI also highlighted that they are making sure that the model rejects text input prompts that violate their usage policies such as prompts that have extreme violence, hateful imagery, sexual content, or the intellectual property of others.