Want to make a video of something that doesn’t exist but don’t have the technical skills to do so? Now, it doesn’t matter — if you can type it, you can see it.
On Thursday, OpenAI revealed its new text-to-video model, Sora, which can generate videos up to a minute long based on any prompt a user enters into a text box, though the technology is not yet available to the public, NBC News reported.
Related: Microsoft Invests Billions in OpenAI, Creator of ChatGPT
OpenAI CEO Sam Altman showed off the technology on X, where he crowdsourced possible prompts and used them to generate several videos, including one of turtles riding bikes across the ocean and another of dogs podcasting on a mountain.
https://t.co/qbj02M4ng8 pic.twitter.com/EvngqF2ZIX— Sam Altman (@sama) February 15, 2024
https://t.co/uCuhUPv51N pic.twitter.com/nej4TIwgaP— Sam Altman (@sama) February 15, 2024
OpenAI said in a blog post that its goal is to train “models that help people solve problems that require real-world interaction,” a prospect that’s exciting for many AI aficionados. But it’s concerning to people wary of the technology’s unchecked power amid the proliferation of deepfaked media of celebrities, politicians and private figures online — and in an election year.
“I am absolutely terrified that this kind of thing will sway a narrowly contested election,” Oren Etzioni, a professor at the University of Washington who specializes in artificial intelligence, told The New York Times. Etzioni is also the founder of True Media, a nonprofit, nonpartisan organization working to identify disinformation online in political campaigns.
Related: ChatGPT Just Got a Game-Changing Update — Here’s What to Know
OpenAI said in a statement that it’s “not making this model broadly available in our products soon” and that it’s developing tools to determine when Sora generates a video and intends to embed metadata to flag the video’s origin, per NBC News.
