What Happened: So, you’ve got the CEO of Netflix, Ted Sarandos, talking about how AI is this amazing new tool that’s going to help people “tell stories better, faster, and in new ways.” It all sounds great, right?
- But here’s the reality: while he’s saying that, OpenAI’s new video-making tool, Sora 2, is being used to flood the internet with some of the ugliest content imaginable.
- We’re talking about a wave of straight-up fatphobic and racist “comedy” videos all over Instagram, YouTube, and TikTok. People are using Sora 2 – which can make unbelievably realistic clips – to create and share videos just to be cruel.
- The examples are just awful. There’s a viral clip, seen almost a million times, of an overweight woman bungee jumping, only to have the bridge “collapse” under her. Another one shows a Black woman “falling through the floor of a KFC,” which is just a disgusting mix of racism and body shaming. Then there are others showing delivery drivers falling through porches or “swelling up” after eating.
- And the scariest part? A lot of people watching these videos think they’re real.
OpenAI
Why Is This Important: This is the big problem with AI that no one wants to talk about. It has put the creation of hate content on steroids.
- What used to take someone with actual production skills a bunch of time to make, any hateful person can now generate in seconds.
- This isn’t just “bad taste” – it’s a real ethical crisis. It’s a way to mass-produce and amplify the most harmful stereotypes about people, all for a cheap laugh.
- It also proves that the “guardrails” these AI companies, like OpenAI, claim its tools have are failing. Miserably. It’s supposed to block this exact kind of hateful, violent content, but it’s clearly not.
Unsplash
Why Should I Care: If you use social media, you’re already seeing this. This isn’t just harmless online trolling.
- This junk shapes how people see the world, especially kids.
- And because so many people can’t tell the difference between what’s AI-generated and what’s real, it completely blurs the line between reality, dark humor, and just pure, unadulterated hate.
- And, of course, when one of these videos goes viral, it just encourages a dozen more people to make their own versions for clicks and “likes.”
What’s Next: So far, OpenAI has been completely silent about this new wave of fatphobic content.
- But this is forcing a much-needed, uncomfortable conversation about who is responsible when these powerful tools are used to hurt and harass people.
- Regulators are definitely starting to pay attention. As these AI tools become as easy to use as a filter, the real challenge is going to be figuring out how we stop all this “creativity” from coming at the cost of our own humanity.
