OpenAI’s new generative AI tool Sora, revealed Thursday, doesn’t have a release date yet, but those itching to play with it can share their text-to-video prompts directly with the company’s CEO, who is posting a select number of results online.
“We’d like to show you what Sora can do,” Sam Altman wrote on X, formerly known as Twitter, after OpenAI announced the product. “Please reply with captions for videos you’d like to see and we’ll start making some.” As of this writing, the tweet has generated around 14,000 responses, including prompt suggestions from bizarre to artsy, serious to snarky.
The new tool can generate videos up to a minute long featuring “highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions” according to OpenAI, which shared an X thread of videos generated by Sora: a scene of a snowy, bustling Tokyo; a papercraft world of a coral reef filled with colorful sea creatures; a short, fluffy little monster kneeling next to a melting red candle, as short, fluffy little monsters do.
Then Altman invited the public to get in on the prompting, adding that people shouldn’t hold back on the difficulty or detail. Wrote one X user who clearly took that directive seriously:
“A herd of four beautiful hyperrealistic racehorses, each riding an astronaut wearing a puffy Balenciaga jacket, atop an avocado-themed saddle, each horse with overlaid text representing the exact names of each of the four Horsemen of the Apocalypse. Set in an empty room with absolutely no elephant in it, no elephant anywhere in the room, dynamic FPV drone shot in 8k, starting wide and then zooming deep to closely the inspect detailed ungloved hands of each astronaut.”
Ready for a bit of weirdness? Here are some of the Sora-generated videos Altman has posted so far based on the public’s suggestions. If you’ve ever wanted to see dolphins and penguins biking across the ocean’s surface or golden retrievers podcasting, this is your lucky day.
Not surprisingly, Altman’s invitation also generated prompts poking fun at OpenAI , particularly the drama that overtook the company last year during a wild five-day November tailspin that saw the surprise ouster of Altman by the firm’s board, followed by an employee uprising over the move and finally, the reinstatement of Altman and president Greg Brockman, who had resigned in solidarity.
Reads one suggested prompt:
“Sam Altman sitting in new hire orientation at Microsoft in November of 2023, scrolling his phone for updates on OpenAI, getting a text message from the board that he’s back, and standing up from his seat and exclaiming, ‘Sammy’s back, baby!” and storming out of the room, to the confusion of the other new hires.”
Generative AI tools are, of course, generating a range of responses, from excitement about creative possibilities to anger about possibly copyright infringement and fear about AI’s impact on the livelihood of creatives—and on creativity itself. Sora is no different.
OpenAI said it needs to complete safety checks before making Sora publicly available. Experts in areas like misinformation, hateful content and bias will be “adversarially” testing the model, the company said in a blog post.
“Despite extensive research and testing, we cannot predict all of the beneficial ways people will use our technology, nor all the ways people will abuse it,” the post said. “That’s why we believe that learning from real-world use is a critical component of creating and releasing increasingly safe AI systems over time.
Read the full article here
