Swarajya Logo

News Brief

'Sora': ChatGPT Maker OpenAI Launches New Text-To-Video Generating Model

Nayan DwivediFeb 16, 2024, 10:26 AM | Updated 10:26 AM IST
Screen Grab from video generated by OpenAi's Sora.

Screen Grab from video generated by OpenAi's Sora.


OpenAI, a leading artificial intelligence (AI) company, has announced its latest innovation: Sora, a cutting-edge video-generation model.

Branded as a tool capable of transforming text instructions into vivid video scenes, Sora represents a significant advancement in AI technology.

Sora can generate videos up to a minute long while maintaining visual quality and adherence to the user’s prompt.

Initially, Sora will be accessible to a select group comprising artists, filmmakers, and researchers focused on identifying potential misuse of AI technologies, known as "red teamers," according to OpenAI's announcement on Thursday (15 February).

Building upon the technology behind OpenAI's DALL-E image-generation tool, Sora interprets user prompts and expands them into detailed instructions.

It then utilizes an AI model trained on video and images to craft the resulting video content.

OpenAI also acknowledged that Sora has weaknesses, including difficulty with continuity and distinguishing left from right.

“For example, a person might take a bite out of a cookie, but afterward, the cookie may not have a bite mark,” the San Francisco-based startup said.

Over the past year, there has been a rapid improvement in the quality of AI-generated images, audio, and video.

Companies such as OpenAI, Google, Meta, and Stable Diffusion have been racing to develop more capable tools.

However, concerns have been raised by democracy advocates and AI researchers regarding the potential misuse of such tools to deceive and manipulate individuals.

Join our WhatsApp channel - no spam, only sharp analysis