OpenAI, the creator of the renowned AI bot ChatGPT, has announced a new video-generating AI, “Sora.” The Sora bot can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions, according to OpenAI.
“Today, Sora is becoming available to red teamers to assess critical areas for harms or risks,” OpenAI says in a press release. “We are also granting access to a number of visual artists, designers, and filmmakers to gain feedback on how to advance the model to be most helpful for creative professionals.”
Also Read: Ripple (XRP) Aims for Massive Breakout from 3-Year Symmetrical Triangle
OpenAI is one of the leading technology companies when it comes to AI aids and content. Its ChatGPT bot is already popular amongst web developers and companies worldwide. Sora, however, expands upon the abilities of ChatGPT, allowing users to auto-generate complex scenes from a prompt. What makes this AI system so powerful and intriguing though, is its ability to understand a prompt as it would pertain to the real world, and adapt to those real-world properties in the scene.
OpenAI’s Sora: A Work In Progress
On the other hand, OpenAI does acknowledge in the release that the model is still a work in progress and has flaws. “It may struggle with accurately simulating the physics of a complex scene, and may not understand specific instances of cause and effect,” the release states. “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.” These issues come with launching a new product, hence why the ChatGPT founder is making Sora available to red teamers to test.
OpenAI is also pledging responsibility in ensuring that people know when their product was used in developing a scene. “We plan to include C2PA metadata in the future if we deploy the model in an OpenAI product.” This tool helps detect misleading content such as a detection classifier that can tell when a video was generated by Sora.
Also Read: Michael Saylor’s MicroStrategy Bitcoin Holdings Worth Over $10B
Like other AI bots, Sora is a diffusion model. This means it generates a video by starting with one that looks like static noise. Then, it gradually transforms the video by removing the noise over many steps. Sora is similar to ChatGPT models, in which it uses a transformer architecture, unlocking superior scaling performance.