This image seems like misleading bull shit. This seems to be the law referenced: https://apnews.com/article/minnesota-artificial-intelligence-nudification-x-elon-musk-deepfake-131184be939d540de093b567b12c9e16
It makes companies liable for people using ai for nudification (which of course includes children). Musk argues the law is a bit to far reaching ( claims the law considers some body parts, that are commonly displayed in public, as private parts and also that the company can get in trouble for people nudifying themselves).
I don’t know how legitimate Musk’s concerns are, but the image is extremely misleading.
And this is why whether it’s Reddit or this site it doesn’t make a difference. Clickbait and dumb people always dominate the conversation and nuance is immediately thrown out the window because nuance does not translate well on the internet when you are trying to gather points.
The worst part is that this guy actually has endless things you can point to to show what a scumbag he is but yet there has to be stuff like this that delegitimizes opposition to him.
Fuck that, he should be 100% liable if he sells a machine that is ised to make child porn wtf you mean clickbait. This is the only way to stop it, otherwise it’s going to happen because the companies will make money off it.
Why not make PC manufacturers responsible for crimes committed on the computer? Because it makes selling computers impossible, the technical requirements are impossible to fulfill in any economic feasible way.
Besides there are open source stable diffusion AI models already that you can train to generate this stuff locally, on your own PC. Do you want all open source AI banned?
Maybe you do. The AI corporations would love that too. I don’t.
I haven’t looked at this in depth, but some of the scholars I follow have suggested that this law is legally problematic, and there are good arguments to strike it down under the first amendment and section 230.
Section 230 is the federal safe harbor law that shields social media from liability for their users content, provided they take certain reasonable steps to moderate the user content. The federal law will preempt this Minnesota state law if they conflict.
But in the case of AI, it’s company a generated content.
So is adobe liable content created or altered in photoshop? You can see AI as an advanced computing platform that comes with a free “worker”. A worker that the user controls and the company can’t really.
