Sora, an OpenAI generative video programjust arrived at Android. According to the company’s announcement, as of November 4, millions of users in the US, Canada, Japan, Korea, Taiwan, Thailand and Vietnam can download the app from Google Play Store and access to tools for creating, sharing and mixing AI-generated videos.
This is an extension that will expand access to digital audio-visual products. However, there is still no date for its arrival in Spanish-speaking countries.
Launch Sora for Android follows the app’s debut on iOS, where it surpassed one million downloads in just five days in September. The app allows users to create videos from text, images, voice and their own face thanks to features that combine language modeling and image generation technology developed by the company.
According to OpenAI’s technical explanation, Sora interprets natural language instructions and converts them into audiovisual sequences, making content creation easy even for those with no prior video editing experience.
The app is powered by an artificial intelligence model called Sora 2.
Among the latest innovations in OpenAI integrated art Soraa standout is the inclusion of so-called “character cameos”: reusable avatars created from a wide variety of objects, from pets to toys or artwork. This feature comes with the ability to connect multiple video clips to create multi-scene clips and leaderboards that highlight the most popular videos and characters in the program.
The company notes in its latest update notes that “once created, each character has its own permissions independent of the user’s personal image: they can be kept only for their creator, shared with general followers, or shared with all Sora users.” It also offers the ability to assign a name and tag to a character, allowing you to mention them whenever you want them to appear in a video.
Since launching the app, OpenAI previewed a new tool that complements an existing feature to create “deep fakes” of users themselves that can be used by other members of the platform when authorized. This same technology now extends to any other selected subject, making it easy to introduce pre-designed characters that can be added directly to the video.
About the creation process, OpenAI explains that cameos can be created using an “original human conceived in Sora”, although it has not yet been specified whether importation of fictitious humans created from other AIs will be allowed, nor has the level of realism adopted for these creations been specified.
The company did not specify how it would distinguish between images of real people and artificially generated figures if any AI-generated character was allowed to be uploaded.
Along with the launch of cameo characters, Sora registration without an invite code is temporarily enabled for users in the United States, Canada, Japan, and Korea, which aims to encourage mass adoption of these new features. According to the company itself, this opening will only be available for a “limited time”.
In parallel with the introduction of these functionalities, the platform faces the beginning of a legal dispute: the social network Cameo recently filed a lawsuit for alleged trademark infringement regarding the use of the term “cameo” in Sora.
Among the technical innovations, the “video stitching” option stands out, which connects several clips to create extensive and complex works, while interactive ratings allow us to determine which videos were the most popular and the characters that are repeated the most.
