Text-to-video conversion AI model is unveiled by OpenAI

A Microsoft-backed startup called OpenAI revealed on Thursday that it is creating software that can generate minute-long videos in response to text directives.

As per the company's statement, the software called Sora can be utilized by visual artists, designers, and filmmakers to obtain feedback on the model and for red teaming, which facilitates the identification of vulnerabilities in the AI system.

In addition to producing numerous shots within a single video, the statement claimed that Sora was able to produce complex scenes with numerous characters, particular motion styles, and accurate details of the subject and background.

As per the company's blog post, Sora has the capability to animate a still image and generate videos by responding to text prompts.

The video creation program follows in the footsteps of OpenAI's ChatGPT chatbot, which caused quite a stir when it was first introduced in late 2022 and could create codes, emails, and poems.

Last year, Meta Platforms, the massive social media company, enhanced its Emu image generation model to include two AI-based features that can edit and produce videos in response to text commands.

The parent company of Facebook also intends to take on Microsoft, Alphabet's Google, and Amazon in the rapidly developing field of generative AI.

According to the company, Sora is still a work in progress and could have trouble following a precise camera trajectory or getting lost in the spatial details of a prompt.

Recently, OpenAI revealed that it was developing tools to identify videos that were created by Sora.