OpenAI has released Sora, its AI video generation tool that can create surprisingly realistic videos out of thin air based on simple textual descriptions.

Like other AI image generators, Sora generates videos from descriptions, but you’re able to also upload a photo or footage of a real person to serve as a basis. In order to prevent potential misuse like deepfakes, however, this feature is restricted to only a subset of OpenAI accounts. “Early feedback from artists indicate that this is a powerful creative tool they value, but given the potential for abuse, we are not initially making it available to all users,” OpenAI said.

Sora provides several features to aid image generation. For example, it can remix footage with text-to-video prompts. Another feature leverages AI to blend two competently different scenes together. There’s also a storyboard feature for generating multiple videos in a row based on a sequence of prompts. All Sora-generated footage comes with C2PA metadata toindicate it’s made with AI, but it’s not clear if that’s foolproof or not.

OpenAI offers two subscriptions you can use with Sora, ChatGPT Plus and ChatGPT Pro. People on the$20/month ChatGPT Plusplan have 1,000 credits to generate up to 50 priority lower-resolution videos (720p) that cannot be longer than five seconds. However, with a ChatGPT Pro plan, you can use Sora to create up to 500 priority videos in the sharper 1080p resolution with 20-second durations. On top of that, ChatGPT Pro lets you download Sora videos without a watermark and generate five AI images in parallel. The only problem is that ChatGPT Pro, available since last week, isquite pricey at $200/month.

Sora wasn’t supposed to drop so early, but OpenAI pushed it live prematurely after artists who had participated in the early alpha testing programintentionally leaked the tool, because the company lured them into “art washing” and used them for “unpaid R&D.” The group’s open letter to OpenAI explained, “We are not against the use of AI technology as a tool for the arts (if we were, we probably wouldn’t have been invited to this program) What we don’t agree with is how this artist program has been rolled out and how the tool is shaping up ahead of a possible public release."