Technology @lemmy.world stopthatgirl7 @kbin.social 4 mo. ago

Disturbing AI Images of Children Found for Sale on Shutterstock

petapixel.com Disturbing AI Images of Children Found for Sale on Shutterstock

Someone used Shutterstock's AI image generator to create them.

24 comments

A note on the page warns, “Shutterstock does not review AI-generated content for compliance with Shutterstock’s content compliance standards.” Adding that users must not generate imagery that is “false, misleading, deceptive, harmful, or violent.”

"Pls don't be bad mmkay?"

"We've done all we possibly can."
I'm curious what they call disturbing, but also don't want to see in case they're right.
- The article doesn't mention nudity but what they described is still pretty fucked.
  
  I think it really depends on what "young girl" means in this context. The title says "children", but nowhere in the article does it say that. So I'm unsure if this is another AI-boogyman article, or something else.
This may be controversial, but I don't care what kind of AI-generated images people create as long as it's obvious they're not reality. Where I worry is the creation of believable false narratives, from explicit deepfakes of real people to completely fictional newsworthy events.
- I agree here. Im not worried about imaginary things except for their ability to appear like actual things and mess with truth.
- I've read that pedophiles are more likely to act out on their urges if they have access to real images. I would guess that this also applies for ai generated images too, even if they don't look 100% real, but I could be wrong on that. Whatever stops them from abusing kids is what I'm for.
  
  I want to say research on the subject has been inconclusive overall. I'd certainly update my view given convincing evidence that fictional images lead to abuse of real children.
  
  Of course, none of that has anything to do with the non-explicit video linked elsewhere in this thread of an adult woman using the toilet.
- so youre okay with generating csam if there is a watermark that says "no children were harmed in the making of this content"?
  
  It's not really CSAM if there is no abuse happening, is it?
There's some pretty weird stuff on there, like kids taking a bath and someone on the toilet: https://www.shutterstock.com/video/clip-26807341-woman-sitting-on-toilet-bathroom-young-girl
- Nope nope nope. Not even risking it.
Called it
In before someone defend pedophiles, oh wait
- Please, the Catholic church are heroes
  
  Edit: wow, didn't think I needed the /s since I was directly linking the Catholic church with pedophilia
Lmao the pics have been removed from the article, anyone got a link?
Why the fuck does these AI know how to generate this shit of children 😕
- These image generating AIs don't need to have been trained on what you want it to output. If you tell it to generate a banana car, it doesn't need to have seen a real banana car before, it just knows what a banana and a car look like, and combines them. Similarly, such AIs would know what naked humans look like and what kids look like.
  
  I see, thanks for the explaination

You've viewed 24 comments.