OpenAI launches Sora 2, an AI video generator that blends creativity with social interaction

OpenAI is stepping deeper into entertainment and social media with the launch of Sora 2, its latest AI model for generating video and audio, paired with a new social platform simply called Sora. Sam Altman, the company’s CEO, described it as the “ChatGPT for creativity” moment, saying the goal is to make it “really easy and fast to go from idea to result.”

From Sora to Sora 2

The original Sora debuted in early 2024 as a proof of concept for AI video. The new version goes much further. Sora 2 produces physically consistent video that follows real-world laws—whether it’s the bounce of a basketball or a stunt involving water buoyancy. It adds synchronized audio for the first time, layering dialogue, sound effects, and ambient noise into the output. That means clips can now feel more like complete productions rather than silent reels.

The system has been trained to follow complex instructions across multiple shots while holding scenes together without distortion. Olympic-level gymnastics, figure skating, or other intricate sequences are within its range.

“We are launching a new app called Sora. This is a combination of a new model called Sora 2, and a new product that makes it easy to create, share, and view videos,” Altman said in a post on X.

Altman admitted the model still makes mistakes but framed it as a step toward something larger: “Creativity could be about to go through a Cambrian explosion, and along with it, the quality of art and entertainment can drastically increase.”

The Sora App

The new app acts as the public interface for the model, launching first on iOS in the U.S. and Canada. It takes cues from TikTok’s feed-driven discovery but pushes users to create as much as consume. Anyone can type a prompt and generate content, remix existing clips, or fine-tune what shows up in their feed based on interest, mood, or time.

One of the most talked-about features is “cameos.” After a quick one-time recording for identity verification, users can drop themselves or friends into scenes, with the model maintaining character consistency across clips. Altman said the team was surprised by how compelling it felt during testing: “The ability to put yourself and your friends into a video—the team worked very hard on character consistency—with the cameo feature is something we have really enjoyed.”

The app is invite-only for now, free to use with limits based on compute availability. OpenAI plans to introduce optional payments during peak demand. ChatGPT Pro subscribers will get access to a higher-quality version called Sora 2 Pro. An Android release and API access are on the way.

Here’s a look inside the Sora app. You can create and remix videos, or even step into the scene yourself with cameos—featuring you and your friends—inside a feed built entirely for Sora content. Watch the demo below.

This is the Sora app, powered by Sora 2.

Inside the app, you can create, remix, and bring yourself or your friends into the scene through cameos—all within a customizable feed designed just for Sora videos.

See inside the Sora app👇 pic.twitter.com/GxzxdNZMYG

— OpenAI (@OpenAI) September 30, 2025

Promise and Risks

Altman was candid about the double-edged nature of releasing a tool like this. He pointed to social media’s mixed legacy, warning of addiction risks and bullying. “It is easy to imagine the degenerate case of AI video generation that ends up with us all being sucked into an RL-optimized slop feed,” he wrote.

To prevent misuse, Sora includes watermarking and metadata for AI-generated clips, parental controls for teens, and human moderators for abuse reports. Safeguards block disturbing content or deepfake misuse, and the company says it will pull the plug if user well-being declines. “The majority of users, looking back on the past 6 months, should feel that their life is better for using Sora than it would have been if they hadn’t. If that’s not the case, we will make significant changes,” Altman explained.

A New Competitor to TikTok?

With Sora, OpenAI steps directly into territory dominated by TikTok and Meta’s video platforms. The draw here is creation-first: less scrolling, more making. Early testers describe the experience as addictive, but in a different way—one that turns anyone into a potential creator.

For Altman, that’s the point. He sees Sora as an opening of the gates. “Even in the very early days of playing with Sora, it’s been striking to many of us how open the playing field suddenly feels,” he said.

How it plays out will depend on whether OpenAI can balance safety and creativity. For now, the launch plants a flag: AI isn’t just generating text or images anymore—it’s coming for video and the social networks built on top of it.

Below is a video of Sora 2 in action.