According to a recent post in the announcement, Openai is currently launching the Sora 2. SORA 2 is the next generation of its AI video and audio generators, promising more realistic, physically accurate and controllable results. Unlike previous models, which are often “deceived” in physics, SORA 2 can generate more reliable simulations.
SORA 2 can create advanced movement videos, such as gymnastics sequences, paddleboard backflips, figure skating triple axels, but it produces synchronized sounds, dialogues and effects. You can also insert people and objects from real videos to accurately reproduce both their appearance and voice.
There is also a new iOS app that launches SORA. This app allows you to create and share AI videos, remix clips of others, and place yourself (or friends) directly into AI-generated scenes using the Cameos feature. If a cameo is used in a remixed clip, it counts as a co-creator (and therefore allows for the power to delete that clip).
Openai emphasizes that the launch focuses on responsible use. Feeds also encourage creation for passive consumption, giving users tools to personalize experiences and minimize addictive scrolling.
At launch, the SORA app will only be available in the US and Canada. SORA 2 is available online. Both require an invitation from Openai.
This article was originally published in the sister publication PCFör Alla, translated and localized from Sweden.
