OpenAI Develops Video Creator Software Based On Text

JAKARTA - Microsoft-backed OpenAI announced on Thursday 15 February that it is developing software that can produce one-minute videos based on text.

The software, called Sora, is currently available for red timing, which helps identify weaknesses in artificial intelligence (AI) systems, as well as for use by visual artists, designers, and filmmakers to get feedback on the model.

"Sora is capable of generating complex scenes with several characters, certain types of movements, and accurate details of the subject and background," the company said in a statement, adding that the software could create multiple shots in one video.

In addition to producing videos from text, Sora can also minimize stationary images, the company said in a blog post. The video maker's software follows OpenAI's ChatGPT, which was released at the end of 2022 and creates a furore around GenAI with its ability to write emails and codes and poetry.

Social media giant Meta Platforms Inc., also strengthened their image generation model, Emu, last year by adding two AI-based features that can edit and produce videos from text. Facebook's parent company also wants to compete with Microsoft, Google, and Amazon in the world of a rapidly transforming generative AI.

Sora is still in the development stage, where the company added that the model may confuse the spatial details of a text and is experiencing difficulties in following certain camera trajectories. OpenAI says it is also developing tools that can distinguish whether a video is produced by Sora.