But was he convicted for it?

The evidence we have points to the opposite conclusion.

After all, in the last 30 years, we've had an absolute explosion in the availability of freely available adult pornography. If more porn access meant more rape and sexual assault, the rates of these should have skyrocketed over the last 30 years. Instead they've gone down. I wouldn't automatically conclude a causal relationship there. But if availability of porn causes rapes and assaults, then we would know that by now.

That's not necessarily true. You could have all sorts of training models that didn't rely on any real children at all. You're just assuming it was trained on actual child porn. Other ways it could be trained:

• Trained on images of adults that look younger than 18. You could commission a bunch of pornography by people with whatever medical condition this guy has and use that as your training data.

• Trained on pornographic images of adults and regular images of kids. (This one is questionable legally, even though again, no kids were harmed in the making of it.)

• Trained on entirely on images of adults, reviewers are tasked with trying to identify the age of the person in the image generated, and using that as training data.

If someone was deliberately trying to create a legal child porn engine (what a horrible phrase to type), they would be sure to do it without using any actual child porn as the training data. There would be little point otherwise.

This is completely wrong. I have no idea how you can say this when it is plainly obvious that ai can create things it never saw.

What? No, that's wrong. They can, for example, create a picture of a monkey made of yarn. Does that mean they were trained on pictures of 'yarn monkeys'?

No, they had pictures of monkeys and pictures of things made of yarn; therefore, they can create pictures of 'yarn monkeys'.

If you want to learn more about that, the specific term is 'latent space'.

If AIs could only reproduce their training data they would be completely useless.

That's assuming it was trained on images of child pornography. There are plenty of ways someone could train such an engine without using any pornographic images of actual children. And if someone was going to go to the trouble of deliberately creating a child porn generator with the specific intention of using it openly and legally? There would be no point in doing so if you had to use illegal images to do it.

who is (or apparently is) a child under 16 years

Again, how do you determine that? Sometimes it's obvious, sure, but sometimes it's not. Who decides the age of a fictional character?

One of the biggest and oldest furry art sites recently went through quite a kerfluffle recently because they tried to strictly implement such a ban ... resulting in mass outrage among users as moderators went around arbitrarily and rather capriciously deciding which characters did and didn't look like they were under 18. Lots of artists had their -- purely well-intentioned -- art deleted and were outraged at being accused of producing "child porn" when they had never intended to. Any characters with small breasts or short height (even if they were otherwise clearly adult) were likely to be targeted, drawing further criticism from body positivity advocates, who pointed out that moderators were de-facto identifying certain body types (which some real people have) as 'not adult' and 'not a real woman'. Artists who did fan art of the Pokemon franchise were particularly hard hit, because at least some of the moderators seemed to be of the opinion that pretty much all pokemon were always underage, and therefore always "child porn". And artists with a more 'cartoony' style tended to get hit a lot harder as well, since moderators appeared to believe that simplified characters with less detail meant they were younger. And then there was the diaper fetish crowd, who lost all their stuff, because moderators decided that any depiction of any character (regardless of any other age indicators) was definitely a child if they were in a diaper.

Thankfully, that ended, and decisively so. The site was reorganized, the policy was heavily rewritten (with user input this time), and the worst offenders among the site staff were all fired. Sexual depictions of minors is still against the rules there, but they instituted much clearer, more specific, and more objective guidelines for determining whether a character is or isn't underage.

And that's just an online art site, where the worst thing that could happen is that your art gets deleted and you get banned! I can't imagine risking prison time based on what some prude with a chip on their shoulder decides about the apparent age of a fictional character.