OpenAI has announced the creation of a new AI model that generates videos based on text descriptions. The Sora model is supposed to create photorealistic scenes from text instructions. And the first samples of generated videos presented really look incredibly alive and realistic.
Sora can generate videos up to a minute long. This AI model understands how objects exist in the world, generates living characters, and accurately interprets props. Sora can also generate video from a still image, and can supplement an existing video with new frames and expand it.
Obviously, this is a new stage in the technological revolution in the field of AI. Sora is still in development, but is already amazing with its capabilities in the same way that Midjourney’s AI once surprised. Currently, select specialists have access to Sora to assess the potential risks and threats posed by such AI. A number of artists, designers and filmmakers will also have access to Sora to provide developers with feedback on how AI can help bring creative ideas to life.
Source:
The Verge