OpenAI announces Sora 2, an AI video and audio app that allows for user ‘cameos’

OpenAI announced a new video and audio generation app in a livestream on Tuesday afternoon. The app, called Sora 2, features photorealistic video creation capabilities and is being billed as a social app for friends to share, remix, and discover AI-generated videos. Sora 2 represents the latest in a wave of multimedia generative AI tools, allowing users to create increasingly realistic images, video and audio. Unlike other generative-AI or social-media apps, Sora 2 allows users to create AI-generated “cameos,” or guest appearances, of themselves and their friends in videos. To create these lifelike cameos, this feature requires users to make a one-time video and audio recording of themselves and verify their identity. Like other social media apps including TikTok, Sora 2 will feature a ‘feed,’ or string of videos tailored to users’ interests. Using a recommendation algorithm, Sora 2 will prioritize videos related to users’ interests, people the user interacts with, or topics the algorithm predicts will then inspire users to create their own videos. Sora 2 builds on the original Sora model released in February 2024. Whereas the original Sora sometimes struggled to represent realistic motion, like a basketball bouncing off of a backboard, OpenAI said Tuesday that Sora 2 “is better about obeying the laws of physics.” The app is available for download now on iOS systems, but access to the service remains invite-only. Users can request access through the app. OpenAI signalled that they may roll out access slowly across the US and Canada, initially giving users “generous limits” on video creation. AI systems like Sora 2 require intense computing power, so companies often have to place limits on user access to ensure the service remains accessible for others. “Transparently, our only current plan is to eventually give users the option to pay some amount to generate an extra video if there’s too much demand relative to available compute,” OpenAI said in its launch announcement, referring to finite computing power. OpenAI also acknowledged concerns about potential downsides from the app, releasing a Sora 2 Safety document in tandem with the larger announcement. Recognizing potential risks of isolation, addiction, and the proliferation of poor-quality AI content, OpenAI said: “we’ve made sure safety is built in from the very start.” To guard against potential risks, OpenAI said they were boosting teams of human moderators to review content for bullying, OpenAI framed Sora 2 as an important step towards more powerful AI systems. In the release statement, OpenAI said: “On the road to general-purpose simulation and AI systems that can function in the physical world, we think people can have a lot of fun with the models we’re building along the way.” “Video models are getting very good, very quickly. General-purpose world simulators and robotic agents will fundamentally reshape society and accelerate the arc of human progress,” OpenAI said.