In a groundbreaking announcement, OpenAI, a leading company in generative artificial intelligence, has unveiled its new model “Sora” that can turn text instructions into highly realistic video scenes featuring multiple characters and specific movements. Founded in 2015 and led by Sam Altman, OpenAI has previously developed ChatGPT and the DALL-E still image generator.
The company has yet to make “Sora” available to the public as they are still analyzing the technology to understand potential dangers and risks. However, OpenAI has shared the technology with a select group of researchers to study its potential malicious uses and develop countermeasures.
Recently, Steven Levy of Wired magazine had the opportunity to witness the “Sora” model in action, although he was not able to use it. When given a text instruction to create a Tokyo city scene on a snowy day with a moving camera, Levy described the result as a compelling and realistic depiction of Tokyo streets. The virtual camera followed characters in the scene, creating an immersive experience that felt lifelike.
OpenAI highlights that “Sora” is capable of creating various types of scenes, from realistic to fantastical, with accurate physics rules applied even in imaginary settings. By offering a range of services in generative artificial intelligence, including text generation and image and video production, OpenAI continues to revolutionize the field.
In conclusion, OpenAI’s “Sora” model represents a significant advancement in artificial intelligence technology with the potential to create highly detailed and immersive video scenes. The company’s commitment to analyzing and addressing potential risks demonstrates their dedication to responsible innovation in this field. With its focus on creating realistic scenes that feel lifelike, “Sora” has the ability to transform how we interact with artificial intelligence technology.