OpenAI has launched Sora 2, its latest video and audio generation model, along with a new social iOS app called Sora.

Sora 2 can generate videos with synchronized audio and allows users to insert themselves or others into created scenes using short video and audio samples.

The company said the model improves on previous versions by better adhering to the laws of physics in generated content, and following multi-step instructions.

Sora 2 is available for free with usage limits, and a higher-quality version will be offered to ChatGPT Pro users.

OpenAI said it plans to add Sora 2 to its API and keep previous models accessible.

Meanwhile, the Sora app lets users create, remix, and share videos, with features that control who can use personal likenesses in generated content.

OpenAI is initially rolling out the app in the US and Canada on an invite basis, with plans to expand to more regions.

The app includes limits and parental controls for teens, as well as content moderation and user privacy features.

🔗 Source: OpenAI

Recent OpenAI developments