Sora shows tech is still moving fast and breaking things
Add Axios as your preferred source to
see more of our stories on Google.
AI-generated video from OpenAI's Sora app. Image: Screenshot via Sora
OpenAI's new Sora app is sparking intense reactions — delighting some users and alarming others who see it as another tool undermining truth and human creativity.
The big picture: Despite the polarized response, the Sora moment is likely to be another step in society's grudging adjustment to AI, akin to past reactions to ChatGPT's image generator and, three years ago, to the debut of the chatbot itself.
Driving the news: The Sora app, released last Tuesday, has risen to the top of Apple's App Store rankings, even as access to generate videos remains by invitation only.
- In a late Friday blog post, Sam Altman said that the company is going to "give rightsholders more granular control over generation of characters," likening it to the way that individuals can choose how their own likeness is used in Sora.
- The company said it needs to monetize video generation.
- That could mean a subscription, ads, or both.
- Altman did not offer timing or further details of how any of this will work, and OpenAI declined to comment beyond the blog post.
Catch up quick: Sora allows people to generate videos that mimic nearly any genre, including copyrighted content from movies and TV shows.
- The app allows people to describe a video they want and insert an AI version of themselves or their friends (with permission).
- Although Sora videos are labeled as AI-generated with both visual watermarks and digital content credentials, many see the app as a supercharger of effortless deepfakes.
- Among the early examples that have gone viral are clips of Sam Altman shoplifting GPUs from Target and another that has Altman watching a series of Pokemon march by, with Altman saying he hopes Nintendo doesn't sue.
What they're saying: Some users have been thoroughly enjoying the viral video tool, while others say it's a glorified slop spreader.
- Creating an instant meme maker seems at odds with Altman's "Abundant Intelligence" blog post last month that said without more compute the company might have to choose between curing cancer and teaching children.
- OpenAI's focus is centered on research efforts and the quest for AGI, Altman responded, but "it is also nice to show people cool new tech/products along the way, make them smile, and hopefully make some money given all that compute need."
Between the lines: By releasing Sora before coming to terms with copyright holders, Altman is allowing creators to see the potential market in action before deciding their approach.
- Altman warned of a bumpy road ahead. "Please expect a very high rate of change from us; it reminds me of the early days of ChatGPT," he wrote. "We will make some good decisions and some missteps, but we will take feedback and try to fix the missteps very quickly."
The bottom line: As with disruptive tech of decades past (Napster, Uber, Airbnb), the "ask forgiveness, not permission" ethos that powered the web's previous waves is back, this time moving at the speed of AI.
