Replying to an earlier post

You can’t have training data for it I guess? But then the model maker would have to break the law to even make it of course so you can’t even use that model without some laws being broken down the line. Then again virtually all models should be illegal because of that. I feel like there is a lot of legislation to be done here because the whole industry is rotten to the core.

Replying to an earlier post

The TL;DR is that there is a voluntary system that tracks court cases and will alert CSAM victims when it suspects images of them (either known images, or new ones) are the topic of criminal investigations. A Jane Doe was repeatedly raped in the early 2000’s to produce copious amounts of CSAM, which was broadly shared among pedophile groups. Her images were apparently very popular and prolific. Doe was alerted that Grok was producing new explicit images of her as a child. The images in question are novel (generated by AI, and not matching any known existing images) but are undoubtedly of Doe as a child. Meaning it was using Doe’s images as part of the training data for the images it generates.

Essentially, if Grok was trained on CSAM, Doe’s popular images were almost certainly included. And now Grok is producing new images of her. It would be like asking Grok to make an image of a popular egirl, and getting an image of what is undoubtedly Belle Delphine. And then X goes “no no no, we didn’t use any images of Belle Delphine in our training data. We promise!

Replying to @⁨Kirp123@lemmy.world⁩

It's important to note that to produce an AI-generated image of somebody, they do not need to exist in the training data beforehand. You can use an image as a prompt, and an AI can create new works based on that prompt, without ever referencing the person from training materials.

Most of these CSAM files are not readily available on the clearnet, and I don't think even Elon is stupid enough to let an AI scraper run free on Tor. One would have to go significantly out of their way to locate clearnet sources of CSAM to include into the training data.

Some would say it's a distinction without difference, but I think it's important to understand how these AI images are actually created, if you want to adequately legislate them.

Replying to @⁨Chozo@fedia.io⁩

Like that article, the actual complaint noticeably leaves some key questions open, while inviting certain conclusions. That these conclusions are not clearly stated as fact makes me fear that the intent is to mislead without lying.

As far as I can tell, the basis for the allegation is that a) CSAM was uploaded to X/Grok and b) the TOS state that uploaded files may be used for training.

Replying to an earlier post

Apparently you can make it and own it if it’s all fake images. Something to do with the legal questions involved, if they illegalized this, then other non perverse things could be illegalized under that precedent. Unless congress steps up, (ha ha,) the judge might actually be in the right here I hate to say it.

Child abuse is always the trojan horse to take away freedoms it seems in the west.

Replying to an earlier post

The same reasoning under our current criminal codes would open up anything and everything the authorities dislike to being forbidden.

Same reason they overturned community moral standard laws a few decades back. Used against the worst cases they could find, but broad enough for them to target everything, like regular pornography. (30 year felony in the precedent setting case, shitholy state GA or something.)

Replying to an earlier post

That is what I read anyway. pedo in chief would presumably quash republican support for a new law, but really with all the dems they could get enough pubs to go along, maybe not to override a veto but enough to force the old baby raper in chief to publicly refuse to sign a law everyone supports, (assuming it’s not a freedom stealing bill, which it absolutely would be now that I think about it.)

Replying to @⁨jpreston2005@lemmy.world⁩

It’s literally just the nazi propaganda playbook for targeting queers, maintained by American conservatives ever since. If americans weren’t so fucking uneducated I wouldn’t need to explain this shit over and over again.

en.wikipedia.org/…/LGBTQ_grooming_conspiracy_theo…

We already have laws against harming children, this shit isn’t about protecting anybody. Any attempt to illegalize a situation where there no actual human victim exists will inevitably be repurposed to target minorities and dissidents. Happened before, about to happen again thanks to countless moralistic morons just like you

Replying to @⁨yesman@lemmy.world⁩

I heard there is a theory from therapists that being able to create that content can help quell urges, and so we should allow it if it protects real kids, vile as it is. However others argue it could cause it cause more to cause harm.

Either way, the decision is well above my ability to decide, and while I would hope actual professionals would have a say, unfortunately it’s going to be a political thing. Pandoras box is wide open though.

Replying to @⁨communism@lemmy.ml⁩

That’s not how AI works. The people claiming that AI models have to be trained on CP to make CP have a fundamental misunderstanding of how image generation models work.

If I ask an AI model to make an image of a sexy air fryer fucking a sexy, anthropomorphic, ear of corn, it will produce said image. I promise it was not trained on sexy images of air fryers fucking sexy anthropomorphic ears of corn.

That’s what makes AI so dangerous and difficult to regulate. We are not ready for this technology to be in the hands of everyone.

Replying to @⁨I_Has_A_Hat@lemmy.world⁩

You’re wrong. Literally from the article

AI-generated CSAM may also present other challenges not raised by the government in this appeal. For example, researchers have found that “a major dataset used for training image-generating AI contained hundreds of CSAM images.” Thus, unlike the virtual CSAM at issue in Free Speech Coalition, the market for AI-generated virtual CSAM images may impact the demand for actual CSAM on which the AI models can train.

Replying to @⁨beep@piefed.world⁩

Wait. Doesn’t the PROTECT Act deal with this?

Prohibits computer-generated child pornography when “(B) such visual depiction is a computer image or computer-generated image that is, or appears virtually indistinguishable from that of a minor engaging in sexually explicit conduct”; (as amended by 1466A for Section 2256(8)(B) of title 18, United States Code).

Wouldn’t this fall under that? I’m confused.

Replying to @⁨beep@piefed.world⁩

Controversial opinion: If there is no real child whose likeness is used to produce the image, then there is no real victim. The art that is created is of the same quality as a drawing from imagination.

The other problem with illegalizing content on the basis of whether it’s CSAM is the eternal question of whether an unreal likeness can be said to even have an age. How would you be able to prove with certainty that an AI-generated likeness was below the age of consent? You would think it would be obvious, but there is a huge gray area here. If the protection weren’t fairly strong, I 100% guarantee that conservative Bible-thumpers would take advantage of the opportunity to ban all ai pornographic content, and then move on to real porn after that.

In addition, Banning AI material we might be able to interpret as CSAM does nothing to protect real children. There is no empirical proof that any kind of porn has a real, independent effect on people’s sexual behavior.

Edit with addendum: In addition, civil rights are a matter of principle. It doesn’t matter whether the presence of AI generated material makes the job of policing more difficult. The Bill of Rights was created to try to make sure that we don’t become slaves. Just because something is highly offensive to my own tastes doesn’t mean that the civil rights of the person who offends me disappear.

Replying to @⁨Mongostein@lemmy.ca⁩

Were they also trained on the vast quantity of media of gymnasts made out of pasta for them to be able to create videos of that?

One of the main points of GenAI is the ability to smush multiple concepts together. There are plenty of training data of children that are in swimwear, or just literally naked that has nothing to do with CP, and there’s boatloads naked adults and of course, porn, it can successfully combine them together even if there was no actual CP in there whatsoever.

www.reddit.comReddit

Replying to @⁨Test_Tickles@lemmy.world⁩

Anything’s possible when you make shit up. You don’t have research to support the claim that viewing any sort of pornographic material leads to behaviors depicted in that material, personal stories and confessions are not even close to the same thing as research and the fact that you bothered listing them together tells me that you’re a fucking moron

Replying to @⁨Test_Tickles@lemmy.world⁩

Most reasonable people would get angry about someone lying about shit and then telling them they’re wrong for calling them out and backhandedly accusing them of being pedophiles for disagreeing, and that is exactly what happens every time this topic comes up. I do have a vested interest in keeping the US from setting up a legal justification/precedent/system for incarcerating and killing trans people, of which I am one, and conservatives have been pushing the narrative that trans people are converting/targeting kids specifically for this reason. Any legal framework targeting pedophiles for anything other than actual crimes against children will be hijacked by conservatives and used against me and people I care about for the crime of being queer.

Replying to @⁨Amnesigenic@lemmy.ml⁩

Repeated exposure and normalization are basic psychology that would be introduced in even the most rudimentary Intro to Psychiatry text book. They are basic tools used by therapists, advertisers, politicians, ect, ect to do everything from good to bad like convince you that spiders aren’t horrifying, black people = white people, being gay is ok, ADHD doesn’t mean you’re defective, that you just should buy certain items because they match the lifestyle you think you deserve, and more recently that being a racist pedo rapist is acceptable as President.

You can claim that you are fighting for gay rights by tossing all the children to the pedos, but that is bullshit. If you are actually gay, then you are purposely equating homosexuality with pedophilia. The only way your argument wouldn’t automatically make you one of the dumbest people on the planet is if you were purposely throwing on the “gay” mantle to reinforce the association of “gay” with “pedo” (like some political huckster), or if you were a pedo trying to deflect people from calling you out as the pedo that obviously are.

Replying to @⁨Test_Tickles@lemmy.world⁩

You’re a complete fucking moron and dead wrong about absolutely everything, your “think of the children” bullshit will inevitably be co-opted by conservatives to target queers, history repeating itself because you’re too fucking stupid to learn from it.

en.wikipedia.org/…/LGBTQ_grooming_conspiracy_theo…

en.wikipedia.orgLGBTQ grooming conspiracy theory - Wikipedia

Replying to @⁨Amnesigenic@lemmy.ml⁩

My dude anything is possible when you gatekeep information. You’re defending making child sexual abuse material (SCAM).

This is what the entire case was sort of about, and also explicitly not about. Does it harm children and the answer was the first amendment. That’s not really an answer. Honestly this isn’t a problem for the courts to figure out it’s a legislative problem and good luck with that.

My personal opinion, no one should be able to create any sexualized content of children regardless of whether or not it’s AI generated, authentic CSAM, hand drawn art, or anything else. Children should not be sexualized, I can’t believe this is up for debate.

Replying to @⁨Amnesigenic@lemmy.ml⁩

Driving drunk is known to be dangerous. People who indulge in that behavior eventually hurt others. We don’t wait for a drunk driver to crash into someone before we arrest them. We arrest them the minute they turn on their car while drunk.

Someone who indulges in sexualizing children will eventually hurt a child and we should not need to wait for it to happen before we arrest them.

Replying to @⁨dawcas@scribe.disroot.org⁩

There is no scientific evidence to support the claim that viewing any kind of porn leads to any behavior other than possibly viewing more porn. People who view furry porn don’t suddenly decide they want to fuck animals, people who view rape fantasy material don’t magically become enthusiastic rapists. People who want to molest children don’t want it because they saw porn about it, they create CSAM as a result of their pre-existing pedophilic urges. Horse, then cart.

Replying to @⁨Amnesigenic@lemmy.ml⁩

I said risk, not that it definitely would happen. Caution, do you know about that? Or reading comprehension.
It’s the same as when a chemical isn’t proven safe and businesses are not allowed to add it to your food. It may be completely safe, but just in case.
If the benefit of using something outweighs the risk it would be reasonable to allow it, but we are talking about something with no proven benefit other than letting pedos enjoy themselves.
Not to talk about the re-victimization of CSAM survivors because there are cases of victims appearing in AI generated material because of the training using the original rape media. But I guess you don’t care and would say they just look alike or some excuse.

Replying to @⁨Amnesigenic@lemmy.ml⁩

from the article:

AI-generated CSAM may also present other challenges not raised by the government in this appeal. For example, researchers have found that “a major dataset used for training image-generating AI contained hundreds of CSAM images.” Thus, unlike the virtual CSAM at issue in Free Speech Coalition, the market for AI-generated virtual CSAM images may impact the demand for actual CSAM on which the AI models can train.

Replying to @⁨aurelar@lemmy.ml⁩

Arguing for the legal production of CSAM is fucked up, even if it’s by an AI.

Also the article specifically states:

AI-generated CSAM may also present other challenges not raised by the government in this appeal. For example, researchers have found that “a major dataset used for training image-generating AI contained hundreds of CSAM images.” Thus, unlike the virtual CSAM at issue in Free Speech Coalition, the market for AI-generated virtual CSAM images may impact the demand for actual CSAM on which the AI models can train.

You’re wrong. We absolutely can draw the line here. And yes, it does harm children. allowing people to create, possess, and distribute CSAM, even if it’s AI-generated, will influence peoples willingness to act on their urges in real life, with real children. It’s not a stretch to consider that linkage. What you’re arguing for is shameful and disgusting.

Replying to @⁨beep@piefed.world⁩

It’s nearly an impossible question to answer in criminal scenario. An actual photo of an exposed minor is clearly harm to that individual. Putting their face on a fake but believable body is nearly as bad. Then you get to the sort of 5-10% gap where it could be them or inspired by them, but few would believe it’s an actual photograph. It all goes downhill from there. Cartoonify someone, give them bigger boobs and put them in a bikini on Mars and we start to wonder how this got to trial. It would be a huge gift to lawyers but I think most of this needs to play out in civil courts.

Replying to @⁨beep@piefed.world⁩

So here’s going to be an unpopular opinion. First let me preface this with I think hurting children is one of the worst things you can do. But is drawing a naked child illegal? Is a video game with a naked child 3d model illegal? Where is the line? Is it weird? Hell fucking yes. Is it wrong? I also believe yes. But it’s not an actual child. The real question should be, how did the AI know how to create those images? Because it was trained on actual child porn. And how this AI was trained should really be the crime. Because it ingested shit-tons of actual CSAM.