OpenAI has recently announced Sora, the company’s latest AI model built for text-to-video diffusion. This joins the current offerings of the generative AI chatbot ChatGPT and the text-to-image model DALL-E.
OpenAI – Sora
The new text-to-video model will generate videos of up to one minute long that incorporate scenes and multiple characters.
The examples of videos showcased by the company include a drone view featuring California during the Gold Rush era and a petri dish with bamboo growing with tiny red pandas.
Currently, Sora is available to red teamers or experts who want to test the models for harm and risks. A select group of visual artists, designers, and filmmakers will also have access to the new model so the company can “gain feedback on how to advance the model to be most helpful for creative professionals”.
OpenAI has clarified that it is building tools to be able to label that these videos were made with the new text-to-video model with C2PA metadata in the future. Moreover, it has the existing safety methods that is applied to DALL-E that allow the model to reject inappropriate or harmful text prompts.
“We’ll be engaging policymakers, educators, and artists around the world to understand their concerns and to identify positive use cases for this new technology,” the company said in a statement.
Moreover, it revealed that it believes “that learning from real-world use is a critical component of creating and releasing increasingly safe AI systems over time.”
You can check out more of the videos on the official website here.
Ram found his love and appreciation for writing in 2015 having started in the gaming and esports sphere for GG Network. He would then transition to focus more on the world of tech which has also began his journey into learning more about this world. That said though, he still has the mentality of "as long as it works" for his personal gadgets.
