AI “undressing”

[morrowine]

Honestly, I’ve been thinking about how fast these AI “undressing” tools evolve, and it’s kind of unsettling. A year ago most of them looked like blurry sketches, and now some outputs already seem way too realistic. I’m wondering where this is all going. Will governments start stepping in and limiting what these tools can do, or will the tech just keep getting sharper until it’s impossible to tell what’s edited and what isn’t? I’m curious how others see it developing, both from a tech and a legal angle.

[sanddylouise]

I’ve been experimenting with different AI photo editors purely out of curiosity, and the jump in realism is definitely noticeable. Even something like Generic Anchor shows how the models are picking up small details that older tools used to miss completely. The tech is clearly moving toward more layered understanding of skin tones, lighting, folds, and all that. But that’s exactly why I think regulation will eventually catch up. Once people start using these tools to create convincing fake images of others without their permission, lawmakers won’t ignore it; they usually step in after a few big scandals hit the news. The funny part is that some developers try adding disclaimers saying outputs are “AI-generated,” but nobody actually reads those disclaimers. It also depends on the country: some places move faster with privacy-related laws, others barely care. So yeah, the realism will grow, but so will the pressure to restrict what’s allowed. Kind of like how deepfake rules started appearing only after celebrities complained publicly.

[kanarycrow12]

Hard to predict anything with certainty, but I agree that technology and regulations usually chase each other in cycles. Tools get better, then someone misuses them, then laws tighten. For now it’s still a gray area, but I imagine in a few years there’ll be clearer limits, especially around using other people’s photos without consent.