Hacker News Reader

Comment by pantalaimon | Hacker News Reader

Maybe I’m naive, but isn’t AI generated CSAM a good outcome, actually - because it doesn’t require actual children to be hurt?

View on HN

This area is already well explored just with fake CSAM generated by artists using photoshop, cartoons, etc. The modern thinking is that it supports and encourages a behavior that can lead to actual violence.

If you constantly watch videos of people eating cheeseburgers, you might want to eat a cheeseburger yourself.

>This area is already well explored just with fake CSAM generated by artists using photoshop, cartoons, etc.

I'm familiar with the research in this area, and that's not something you can say confidently; most work (and by that I mean 2 or 3 papers in total) has gone into investigating the role 'generated' depictions of CSAM play in the collections of hoarders. No psychological study, as far as I'm aware, has conducted an investigation on those who enjoy cartoon material akin to what you might find in a Japanese manga.

In fact, there's some evidence against what you're saying; anthropological research on fans of cartoon material ('lolicons' or 'shotacons') in Japan shows that their communities draw hard lines between '2D' and '3D' not just in this area of sexuality, but in their sexualities as a whole. This sexual inclination toward the 2D world is termed the 2D-complex and is akin to 'digital sexuality' or fictophilia, not pedophilia.

By way of analogy, perhaps BDSM would work as a good counter point to you. Many people (some studies suggest the majority of people) engage in 'rape fantasies' or other such fantasies of illegal or immoral nature, yet although actual depiction of rape is rightly banned by the state, its simulated variants are not, and we are comfortable to acknowledge that sexual desires do not always manifest in real life, and sometimes the thrill of fantasy itself is the attraction. To make it real would, ironically, defeat the whole point.

One issue with this is that the fake CSAM wouldn't be the cartoons of a Japanese Manga, it would (or could) be photo realistic. It could be photo realistic of real children. This is obviously possibly bad because it might fuel or encourage pedophiles, but it also has lots of other negative possibilities too.

One example of a bad thing - you could easily imagine an instagram bot that looks for pictures of people with their kids, then uses a Stable Diffusion like model to produce pictures of the people having sex with their kids, or horrible things happening to the kids, and reply to the target account. The bot might threaten to post the pictures and accuse the person of being a pedophile unless the person pays X in bitcoin (or whatever). Or, the bot could just post such pictures for fun.

I think we don't know if fake CSAM will have a good or bad effect on pedophiles and, sadly, there is no real way to reliably test that (so far as I know). Fake CSAM might placate pedophiles, or it might whet their appetite. It's hard to know what to do.

I think we will eventually get to the point where very good unrestricted image generation models are available to the general public. When that happens there will be chaos - you will live to see man-made horrors beyond your comprehension.

That's a good point, and I agree - I only wanted to pick up on the point about 'cartoons'. As for whether realistic generated images with real human data sources would have a good effect, it certainly wouldn't have a good effect on preventing further child abuse, and again, as far as I know there's no evidence that it would 'placate' them in the sense of the (widely debunked) catharsis theory.

And of course, I'm not looking forward to the world ushered in by free roam with this technology, mainly for the reasons you stated.

>I'm familiar with the research in this area, and that's not something you can say confidently

I’m referring to legally, sorry I should have specified.

> The modern thinking is that it supports and encourages a behavior that can lead to actual violence.

Hasn't this nonsense been thoroughly debunked by multiple studies at this point? I would assume evidence and "modern thinking" supports the exact opposite of what you claim, unless by modern thinking you mean the same thinking that tries to hide research they don't like.

Video games do not cause violence. End of story.

The pleasure centers activated by videogames and pornography are quite radically different; I would not assume that the reactions to simulated sexuality is the same as simulated violence.

Then ban porn especially skits that depict actions that society deems deplorable like suffocation and rape, or are the pleasure centers for those also different.

> I would not assume that the reactions to simulated sexuality is the same as simulated violence.

A would not assume anything. Conduct research and draw conclusions. Don't speculate.

People who had no clue what videogames were, were the ones arguing that playing a violent video game would make you want to commit actual violence. The counterargument that players made was that they could "tell reality from fiction" - i.e. that when they played Mortal Kombat or Call of Duty, they put their "Real Life" brain away and put on their "Fictional Video Game" brain, so videogames can't make people violent.

This is the right conclusion, but the logic is entirely wrong.

The reason why video games do not cause violence is that play violence is not anywhere close to the real thing, not that people firewall off fiction from reality. There's plenty of cases in which a piece of fiction has changed people's views! Crime shows are notorious for skewing how actual juries rule on cases. Perry Mason[0] taught them to expect dramatic confessions and CSI[1] taught them to weigh whiz-bang forensics over other kinds of evidence.

In the specific case of porn, there isn't really a difference between "play sex" and "real sex": they poke the same regions of your brain. And the people who are responsible for keeping actual pedophiles from reoffending are pretty much unanimous that the worst thing you can do is give them a bunch of, uh... let's call it "material". So if you're already a pedophile, giving you access to simulated CSAM won't substitute for the real thing. It'll just desensitize you to reoffending.

[0] https://en.wikipedia.org/wiki/Perry_Mason_syndrome

[1] https://en.wikipedia.org/wiki/CSI_effect

A lot of claims and no supporting research. My position is clear: You need to give clear evidence that X causes harmful Y before we can discuss banning X. We don't ban X because you and I find it deplorable.

>> Conduct research and draw conclusions. Don't speculate.

Just like how the incredible availability of porn on the internet has led to millennials being the generation that has the most sex ever: https://news.ycombinator.com/item?id=12433236

Correlation

There are way too many factors at play to simply point at porn, which is probably harder to obtain now in all honesty. I found many random porn magazines/pages as a child. Never did I ever go looking for it, but finding it was always a thrill.

People buy less magazines now (based on convenience store shelves increasingly excluding them.)

You think porn is harder to obtain with the internet? That seems... unlikely.

Modern thinking doesn't mean evidence-based thinking. To the contrary, it gets even more politicized, rather than becoming more evidence-based. Here's an evidence-based counterargument [0].

[0] Evidence Mounts: More Porn, Less Sexual Assault. https://www.psychologytoday.com/us/blog/all-about-sex/201601...

This is like claiming video games cause violence, which is absolutely not the case.

More likely people will just generate more synthetic content to consume.

IMO it's more like claiming that video games lead to video game fans and addicts. Which is true.

Yeah, people who like looking at synthetic images will have easy access to more synthetic images (and could even generate them on their own machines).

But they are not harming anyone else.

...but they're becoming child porn addicts

>If you constantly watch videos of people eating cheeseburgers, you might want to eat a cheeseburger yourself.

this is retarded, if you watch a movie, play a video game, read a book with crime events then you will become a criminal. We have a ton of shooter games and still no evidence that this caused more gun violence around the world.

> This area is already well explored

It's not well explored at all and you just made that up lmao.

That's akin to the idiotic arguments of the past that "allowing people to see homosexuality will make them homosexual!"

Completely ridiculous.

And having a lot of LGBTQ friends leads one to become LGBTQ?

So if I start watching gay porn, I can become gay (or at least bi)? Why don't more people do this and double their dating pool?

If you are not gay you won't enjoy the porn and it will impact you different.

Same with CP. You have to be sick to enjoy it. Very sick.

Then why is GTA 5 legal? Or Hannibal? Is it ever possible to trust anyone with self-determination?

Which is why, after playing so many RPGs, I've become a sword-swinging serial killer. /s

That's what they said about video games. We know how that played out.

If a boy constantly watch media of pretty girl, he may want to become a pretty girl himself. Which is fine IMO but traditional parents are not worried about this possibility much.

Shouldn't call of duty and game of thrones be illegal then?

So long as the AI can remain creative. Once it has exhausted itself of that while the niche consumers still crave for more variation. That's when children start to get hurt again.

Reply on news.ycombinator.com