cross-posted from: https://programming.dev/post/3974080
Hey everyone. I made a casual survey to see if people can tell the difference between human-made and AI generated art. Any responses would be appreciated, I’m curious to see how accurately people can tell the difference (especially those familiar with AI image generation)
A couple were surprising but others seemed obvious in hindsight. Some of these AI models have a really specific vibe that is easy to spot. It can be removed sometimes but if the prompts don’t prevent it, the images tend to have this glow and pop that many real images don’t have. They’re perfectly detailed if that makes sense.
Got 14/20, which I feel pretty good about, but you do this survey every year and it’s gonna keep going lower. I bet even a year ago, most people would be above 75% accuracy.