OpenAI Releases New Sora-Generated Videos: They’re Mind-Blowing!

© OpenAI

While it already offers ChatGPT and the DALL-E 3 artificial intelligence model, which generates images from text, OpenAI has also developed a technology called Sora that generates videos. As a reminder, Sora was presented in February. And if this technology is still not offered to the general public, OpenAI collaborates with some artists, who have used this artificial intelligence model to generate video clips. Regularly, OpenAI's YouTube channel publishes videos called “Sora Showcase” that show what AI is capable of when it is in the hands of creators.

OpenAI recently released two new videos generated with Sora, which are stunningly realistic (although there are still plenty of inconsistencies). One of these videos was with Sora by London-based artist David Sheldrick. It’s a look back at the beginning of his career, when he was photographing behind the scenes at fashion shows. Regarding the technology, he explains: “I’ve used a lot of different AI tools, but nothing comes close to the realism of Sora.”

Subscribe to Presse-citron

The other video is by Niceaunties, an artist based in Singapore. His video is “an art project about aging, beauty, freedom and pleasure, and an attempt to understand the culture of aunties.”

200% Deposit Bonus up to €3,000 180% First Deposit Bonus up to $20,000

No release date has been announced yet

While the Sora model has huge potential for artists, it could also be dangerous if it falls into the wrong hands. “We will take several important security measures before making Sora available in OpenAI products. We are working with red teams—experts in areas such as misinformation, hate speech, and bias—who will adversarially test the model,” ChatGPT’s creator said in February.

It is also developing technology that will identify videos generated with Sora and says it is using other techniques to identify misuse that violates its policies. No release date has been announced by OpenAI at this time. But our colleagues at Tom's Guide believe that OpenAI will not release Sora before the end of the presidential election in the United States.

• In addition to ChatGPT and the DALL-E 3 image generator, OpenAI has also developed a technology called Sora that is capable of generating videos
• For the moment, this AI is not available to the general public, but OpenAI regularly publishes content that has been created with its technology
• Sora could be a powerful tool for artists, but OpenAI must first ensure that this model, capable of generating ultra-realistic models, is not used for bad purposes

📍 To not miss any Presse-citron news, follow us on Google News and WhatsApp.

[ ]