26 Feb 2024
Sora uses your texts to understand your video requirements. It can create videos of varying lengths, sizes, and shapes.
OpenAI launched a text-to-video generator, Sora, in early February. It uses a diffusion model to break down your text input and create a video. It can generate complex scenes with multiple characters, varied backgrounds, and motions. The tool shows an advanced understanding of how things exist naturally and their physical behaviour. It can also understand human language and create characters showing specified emotions.
Sam Altman, the CEO of OpenAI, invited several social media professionals to try out the new app, and the responses were ecstatic. The videos seemed real-life-like. But this also raised several issues about the societal implications of this technology.
Currently, Sora is mainly available for Red Teamers, the cybersecurity professionals who act as enemies to assess the associated risks. Many designers, filmmakers, and visual artists are also given access to try out Sora. Meta and Google are working on similar tools.