OpenAI Premieres Amazing Videos Made With New Sora Text-to-Video Generative AI Model
OpenAI has introduced a new generative AI model named Sora that can generate videos up to one minute long from text prompts. The company described their vision of simulating movement in the physical world to solve real-world problems, and the video demonstrations are an impressive debut for the model.
Sora Video
While not the first of its kind—with examples like Google’s VideoPoet, Meta’s Emu, and startups like Pika and Runway collection already demonstrating similar technology—Sora has captured attention for the high quality of its outputs. Its ability to accurately interpret prompts and create videos that adhere closely to user descriptions sets it apart. The model can also grasp how to express emotions specified in the prompt for characters that are then expressed visually. Sora is also able to render multiple camera angle shots within a single generated video. The company boasted of Sora’s capabilities even as it cautioned about its imperfections.
“Sora is able to generate complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background. The model understands not only what the user has asked for in the prompt, but also how those things exist in the physical world. The model has a deep understanding of language, enabling it to accurately interpret prompts and generate compelling characters that express vibrant emotions,” OpenAI wrote in its announcement. “The current model has weaknesses. It may struggle with accurately simulating the physics of a complex scene, and may not understand specific instances of cause and effect. For example, a person might take a bite out of a cookie, but afterward, the cookie may not have a bite mark. The model may also confuse spatial details of a prompt, for example, mixing up left and right, and may struggle with precise descriptions of events that take place over time, like following a specific camera trajectory.”
Access is being granted to artists, designers, and filmmakers for feedback on potential creative use cases. Security researchers are also testing Sora for potential harm.OpenAI says it is sharing research progress early to gather diverse external input into how the technology can be advanced responsibly. Text-to-video generation has raised concerns over potential misuse.
Sora can also create videos that mimic animation and computer-generated characters that might be used in a film or even for a video game. Generative AI for creating digital assets for games and films has been on the rise anyway, but Sora might skip the middleman of designing assets and go straight to animating them, as seen below.
Follow @voicebotaiFollow @erichschwartz
Google Shows Off Flexible New Generative AI Video Model VideoPoet
