In the Bling Zoo, a tiger wears a giant gold medallion, a monkey sports a bejeweled crown, and a turtle munches on a bowl of diamonds.
Unfortunately, this fantastical destination does not exist. It’s a creation of Sora, the new text-to-video artificial intelligence program from ChatGPT maker OpenAI.
‘Bling Zoo’ was just one of a series of videos Sora created Thursday when CEO Sam Altman asked his followers on X (formerly Twitter) to submit commands that were generated into movies.
The results were so ultra-realistic, they led one observer to comment: ‘This one convinced me the future is here and it’s going to be OK.’
One user requested that Sora create, ‘An instructional cooking session for homemade gnocchi hosted by a grandmother social media influencer set in a rustic Tuscan country kitchen with cinematic lighting’
This prompt led to the most realistic video containing a human that Altman posted on Thursday. Users marveled at how realistic the woman’s hands were – a notoriously tough subject for AI images to recreate
Altman started the stunt with a tweet, saying: ‘We’d like to show you what Sora can do, please reply with captions for videos you’d like to see and we’ll start making some!’ he wrote.
‘Don’t hold back on the detail or difficulty!’ he added in a follow-up post.
The prompts quickly started coming in:
‘A wizard wearing a pointed hat and a blue robe with white stars casting a spell that shoots lightning from his hand and holding an old tome in his other hand,’ wrote one replier.
‘A half duck half dragon flies through a beautiful sunset with a hamster dressed in adventure gear on its back,’ wrote another.
Altman delivered the results, posting some of Sora’s creations, compiled in the following video:
One observer likened it to Interdimensional Cable, an episode of the science fiction TV show Rick and Morty where a special cable box allowed viewers to glimpse television in alternate realities – a world where everyone is made of corn, for example.
‘Bling Zoo’ and other videos like it were not far off from that.
In response to Sora’s video of ‘A bicycle race on ocean with different animals as athletes riding the bicycles with drone camera view,’ Sora delivered a video that led one commenter to speculate about Sora’s supremacy over Dall-E, one of the existing generative AI art programs:
‘I have feeling that any one frame grab from Sora is better than Dalle,’ they wrote.
Sora created short videos of user prompts submitted over X, leading some to liken the results to a science fiction view into an alternate universe
The results were eerily realistic
‘A bicycle race on ocean with different animals as athletes riding the bicycles with drone camera view,’ requested one follower.
And Sora delivered.
Same when someone asked Altman to have Sora show ‘Two golden retrievers podcasting on top of a mountain.’
For one video, a commenter requested
The resulting video showed just that, and the AI-generated woman even waved her hand around to show that she had normal fingers – which can be notoriously hard for AI. Often AI-generated people end up with too many or too few fingers.
‘Big flex waving those fingers around in slow-mo!’ one commenter wrote. ‘And there’s only 10!’ replied another.
Sora is initially launching to select creators, Altman wrote on X. He and OpenAI have not announced when it will be released to the general public.
One particularly impressed X user concluded that ‘AGI is here’.
When an X user requested ‘Two golden retrievers podcasting on top of a mountain,’ Sora delivered
The abbreviation stands for artificial general intelligence, an AI system that can operate on its own without human control, can understand itself, and can learn new skills.
Such a system could be capable of solving complex math or scientific problems that might take humans years to unravel, some scientists hope.
AGI is both a goal and a fear among computer scientists who work with AI, though, as some worry that such a system could view humans as a threat to be eliminated.
In the case of Sora, the computing power is not going to destroying humanity, but to creating 10-second videos from people’s playful prompts.
