I can find no proof of this in the articles you linked.
> You can't do that without fine-tuning on actual images of children.
You definitely can. That's like saying "you can't generate images of a dog with wings without fine-tuning on actual images of dogs with wings".
Any AI image generation algorithm generates realistic, life-like images. I'm not going to prove my point by generating this shit, there's not a doubt in my mind that you can generate these images without any fine-tuning.
Nobody is denying that, but that is not the only way to make it better at this task.
> seems like by allowing this you could be unintentionally creating a way to launder real-life CSAM through generative AI.
There's no way to launder real-life CSAM. If there's a victim, it's a crime, regardless of whether the victim's images went through a ML model or not.
The point is with an AI-generated image you don't necessarily know whether there's a victim or not. You can take an image that wouldn't exist without training data sourced from countless victims, and credibly claim everything is perfectly legal and that there wasn't any victim. That's the danger I'm referring to.
Spending hundreds of thousands training an AI model to generate CSAM sounds like a terrible way to smuggle it across the world. You could feasibly add hidden activation keywords to enable CSAM generation mode without the public noticing, but you could also just put an encrypted file in a model exported as .pickle and it would take years before anyone notices.
And because these models would be naturally superior to models not trained on the real thing (as you yourself just admitted) then they would outcompete those models, thus indirectly funding real-life CSAM.
2) There are adult porn actors who resemble children (and many of those build their porn career on that), you can fine tune on their images.
There are guides on the regular net that teach you how to fine tune on anything, and there is nothing at all that would need to be added or altered in those instructions for fine tuning on CSAM material. I'll take your word for the results of your searches of the darknet for CSAM model weights; I've never done that, as I have no need for CSAM model weights or, really, any other reason to explore the dark net.
> You need to fine tune generative models on actual CSAM to get the most realistic looking images.
I’m not going to spend time trying to create simulated CSAM images to test this, but my experience with using Stable Diffusion in oother contexts says that’s probably not true, though its probably (ignoring the issue of sourcing training imagery) the obvious and straightforward way. But the ability of models to generalize and combine concepts means you don’t generally don’t need to fine tune on the specific combination of concepts you are generating images of. A general purpose model can often produce photorealistic representations of concepts its not trained on when conditioned with a LoRa trained purely on, say, anime representations of the concept.
> The prosecutors noted how lifelike and realistic the images were
Nothing in those articles refers to the darknet, CSAM-specific models, or fine tuning on CSAM imagery. It just says the guy was arrested for using an AI generator to produce imagery that resembled CSAM imagery that was never distributed, under a legal framework which had also allowed that for sexually explicit animations of fantasy subjects that resemble adolescents, and that the prosecuting authorities claim that the imagery was particularly lifelike. As support for your claim of requiring actual CSAM to fine tune to produce realistic CSAM-resembling imagery, it is perfectly circular, resting entirely for its essential premise on the conclusion it is offered to support.
> You can’t do that without fine-tuning on actual images of children.
Not all actual images of children are CSAM, and most models are trained on actual (non-CSAM) images of children to start with. Heck, there are common recommendations for prompt patterns to steer clear of generation with SD1.x-derived models (which probably apply to other models, excluding the ones that are so aggressively self-censored as to make it difficult to even intentionally produce nude imagery of adults, like the SD2.x base models) to avoid producing conditions in which models may combine concepts to produce CSAM-resembling images without user intent. [0]
[0] common ones including avoid “young”, “girl”, or “boy” (the latter outside of the established pattern on which many models are trained of using 1girl/1boy/2girls/etc. to identify number and gender of subjects without regard to age) with prompt terms calling for or leaning toward nudity, sexualized clothing or appearance, poses or other compositional features which would take on an inappropriate character with children involved, and, a fortiori, explicit sexual concepts.
https://news.koreaherald.com/view.php?ud=20230925000652 https://thereadable.co/korean-man-charged-for-creating-ai-ge...
You can't do that without fine-tuning on actual images of children.