OpenAI’s Sora: The Whiz Kid AI Transforming Simple Prompts into Mind-Blowing Photorealistic Videos

“OpenAI’s Sora Turns AI Prompts Into Photorealistic Videos”

“In a demo for reporters on Tuesday, the company showed off an AI character model that can operate robots, play chess, or generate text. The AI, named Dactyl, is controlled by a new, more powerful edition of a language-based model known as GPT-4, which OpenAI has christened Sora.”

In the sphere of artificial intelligence, OpenAI is back at it again, making waves with its grand showcase of a revolutionary AI character model named Dactyl. This tech whizz, not restricted to just tickling the ivories of a piano, is proficient in running robots, showing off some serious chess moves, and – for a touch of the Renaissance spirit – generating text. But hang on a second, don’t mistake it for the run-of-the-mill AIs we come across every day. No siree! Dactyl is powered by the latest, beefed-up version of a language-based model, GPT-4. Bored of the dry, number-based moniker? Don’t worry – OpenAI has given it the much cooler name of ‘Sora’.

Continuing on its roll, OpenAI illustrated how Dactyl can persuasively simulate picking up handy objects and manipulating them. To astound you all the more, this fantastic foray into the world of the future doesn’t stop at digital simulations. Sora – yes, the upgraded GPT-4 – can apply its talent to real-life physiognomies and pairs with a physical robot, which OpenAI has charmingly called a ‘Shadowhand’. Now, isn’t it intriguing to see AI moving from the screens to our living rooms and workspaces?

What’s more? Apart from helping eager-to-learn AIs understand their place in the world – slight exaggeration, maybe? – the researchers at OpenAI are trying to level up the interaction between humans and AI. And surprise, surprise, it involves a text conversation interface, because we so didn’t see that coming! The idea appears to be that users speak or type their intentions, which the AI then fulfills – whether that’s composing a beautiful sonnet, or, well, squeaking out a groovy chess move.

While the OpenAI folks must be congratulated on such tremendous achievements, the plight of the rest of us, who are grappling daily with preserving human relationships, must not be forgotten. Undeniably, the demonstration of such advanced and multi-faceted AI interactions heightens the tantalizing promise of AI and the impending transformation of our lifestyle. But, as we move towards unlocking the full potential of AI with Sora, let’s not forget Dactyl’s impressive endeavor to maneuver our world – like learning not to drop the can of soda!

Read the original article here: https://www.wired.com/story/openai-sora-generative-ai-video/