While doomscrolling today, I came across the news of Sora 2’s release. For those unaware, it’s OpenAI’s text-to-video model. I saw some of the videos people have been generating, and I can say we have come a long way from the original model. It is able to create pretty realistic videos that are hard to distinguish from real ones—until properly examined, which I doubt most people on the internet actually do while doomscrolling.

We have entered a new age of disinformation. People can churn out racist content, deepfakes of celebrities, and bullying could hit a whole new level. What about elections? I doubt a rural person in a village somewhere would really think that a video of a politician saying something is AI-generated.

Worse yet, the internet is about to be flooded with algorithm-bait garbage. In a world where attention spans are already shrinking, this is yet another digital demon we have to contend with.

“A lie gets halfway around the world before the truth has a chance to get its pants on”