OpenAI releases Sora video generator to ChatGPT Plus and Pro users

In short: OpenAI has publicly released Sora, its tool for making short videos from text, to ChatGPT Plus and ChatGPT Pro users in the US and Canada.

What happened

OpenAI first previewed Sora in February 2024. The company then made its first public release in December 2024, but only for paying ChatGPT subscribers on the Plus and Pro plans in the US and Canada.

Sora is a “text-to-video” AI tool, which means you can type a description and it will try to generate a video that matches it. OpenAI says Sora can create videos up to 60 seconds long, and it can include detailed scenes, camera movement (like a camera panning across a room), and multiple characters.

Sora can also animate a still image, or extend an existing video. Under the hood, it uses a method called “diffusion,” which is like starting with TV static and slowly turning the static into a clear picture, frame by frame.

Why it matters

Tools like Sora could change how everyday people and small teams make video content, since typing a prompt can be faster than filming and editing. At the same time, as these tools spread, it can become harder to tell what videos are real and what videos were generated by a computer, which matters for trust online.

Source: The Verge AI