News for hackers. Stuff that matters.
Re:Stable Diffusion 2.0 Release by datalopers on Wednesday November 23, 2022 @09:52PM (#33727344)
They’ve ensured the only way to create CSAM is through old-fashioned child exploitation, meanwhile all perfectly humane art and photography is at risk of AI replacement.

This is a huge missed opportunity to actually help society.

Re:Stable Diffusion 2.0 Release by braingenious on Wednesday November 23, 2022 @10:40PM (#33727579)
LMFAO

What do you propose? The FBI releases a CSAM data set for devs to use for “training”?

Would you be the one to create the model? Would you run a business that sells synthetic CSAM?

Re:Stable Diffusion 2.0 Release by datalopers on Wednesday November 23, 2022 @10:53PM (#33727634)
Without the changes they made to Stable Diffusion, it was already able to generate CP. That's why they restricted it from doing so. It did not have child pornography in the training set, but it did have plenty of normal adult nudity, adult pornography, and plenty of fully clothed children, and was able to extrapolate.

Anyway, one obvious application: FBI could run a darknet honeypot site selling AI-generated child porn. Eliminate the actual problem without endangering children.

Re:Stable Diffusion 2.0 Release by tick_tock_tick on Thursday November 24, 2022 @02:59AM (#33728828)
> FBI could run a darknet honeypot site selling AI-generated child porn. Eliminate the actual problem without endangering children.

It's very unlikely AI generated child porn would even be illegal. Drawn or photoshopped photos aren't so I don't think AI generated would be.

Re:Stable Diffusion 2.0 Release by barrkel on Thursday November 24, 2022 @03:35AM (#33729007)
This isn't the case in law in many countries. Whether an image is illegal or not does not solely depend on the means of production; if the images are realistic, then they are often illegal.

https://en.m.wikipedia.org/wiki/Legal_status_of_fictional_po...

Don't forget that pornographic images and videos featuring children may be used for grooming purposes, socializing children into the idea of sexual abuse. There's a legitimate social purpose in limiting their production.

Re:Stable Diffusion 2.0 Release by snek_case on Wednesday November 23, 2022 @11:06PM (#33727681)
Stable diffusion is able to draw images of bears wearing spacesuits and penguins playing golf. I don't think it actually needs that kind of input to generate it. It's clearly able to generalize outside of the training set. So... Seems it should be possible to generate that kind of data without people being harmed.

That being said, this is a question for sociologists/psychologists IMO. Would giving people with these kinds of tendencies that kind of material make them more or less likely to cause harm? Is there a way to answer that question without harming anybody?

In the mean time, stay away from 4chan.

Re:Stable Diffusion 2.0 Release by sroussey on Wednesday November 23, 2022 @11:28PM (#33727794)
Well, Microsoft and others have this model for recognizing CSAM, trained on those CSAM images.
Re:Stable Diffusion 2.0 Release by PartiallyTyped on Thursday November 24, 2022 @01:26AM (#33728325)
Apple, and meta have as well.

Apparently Facebook has a huge problem with distribution through messenger.

Re:Stable Diffusion 2.0 Release by bambataa on Thursday November 24, 2022 @03:03AM (#33728844)
Once I read an article about a guy who got arrested because he’d put child porn on his Dropbox. I had assumed he’d been caught by some more sophisticated means and that was just the public story. I’m amazed that anyone would be stupid enough to distribute CSAM through an account linked to their own name.
Re:Stable Diffusion 2.0 Release by boppo1 on Thursday November 24, 2022 @04:05PM (#33736218)
I imagine the problem with messenger is teenagers sexting each other.
Re:Stable Diffusion 2.0 Release by PartiallyTyped on Thursday November 24, 2022 @04:42PM (#33736560)
You will find very few teenagers in messenger, most use snapchat instead.
Re:Stable Diffusion 2.0 Release by concordDance on Thursday November 24, 2022 @03:49AM (#33729076)
Yes to the first and no to the second seem the obvious answers here.
Re:Stable Diffusion 2.0 Release by braingenious on Thursday November 24, 2022 @04:52AM (#33729485)
After the FBI releases a database of CSAM to the public for training purposes (the thing you said yes to), who do you suggest are the parties that build the model and run the synthetic csam company?

Since you answered yes to question one, here’s the final question: how do you clean up the release of the biggest CSAM collection in existence from becoming a mainstay on the HD of every pedophile on the planet?

This stuff is so hilarious, like how is it not hard to arrive at “maybe don’t distribute csam” as a basic ending point of thinking? How does somebody that doesn’t want to look at csam think that it’s important that they should have access to it?

Re:Stable Diffusion 2.0 Release by concordDance on Thursday November 24, 2022 @07:31AM (#33730571)
So your hypothesis is that if the FBI gives the database to a company it will inevitably leak to the pedophile underworld?

I can't judge how likely that is.

I guess I also don't care much as I only really care aboit stopping production using real children, simulated CSAM gets a shrug and even use of old CSAM only gets a frown.

Re:Stable Diffusion 2.0 Release by braingenious on Thursday November 24, 2022 @03:31PM (#33735887)
What company? How is it that people are advocating for the release of this database yet nobody says to whom?

My (lol now flagged) opinion is that it’s kind of weird to advocate for the CSAM archive to move into [literally any private company?] to turn it into some sort of public good based on… frowns?

Re:Stable Diffusion 2.0 Release by alignItems on Thursday November 24, 2022 @03:34AM (#33729001)
CSAM is a canary for general AI safety. If we can’t prevent SD from creating CP, will we be able to stop robots from killing people?
Re:Stable Diffusion 2.0 Release by npteljes on Thursday November 24, 2022 @04:20AM (#33729271)
I don't think, and Stability's CEO also doesn't seem to think, that society would receive it as a benefit. Therefore, it's undesirable, right now.