VIDEO: OpenAI Introduces Sora Neural Network for Video Generation

The creators of ChatGPT and Dall-e, OpenAI, have introduced a new neural network named Sora, capable of generating realistic videos based on user text prompts, as reported by rbc.ru. This development has the potential to change the approach to creating visual content.

Sora allows users to create videos lasting up to 60 seconds while maintaining high image quality and accurately following user instructions. The main goal of the project is to train artificial intelligence to understand and model the physical world in motion.

Sora can produce videos that include realistic scenes of both people and animals, as well as clearly fantastical settings, such as mammoths wandering through a snowy field or floating points in the air between buildings.

Currently, the model is being made available to information security experts to assess potential threats and risks, as well as to artists and designers to gather feedback for further improvements to the model.