Users don't/can't read them and service providers know it. There's no actual agreement because anything (kid, bot, dog) could click the button.

If I assume the story is true, then this bit is just some outlets attempt to polarize and trigger.

It's a fucking lawsuit and OpenAI has to say a bunch of shit that is just fact.

I am not going to dig into the kid's life but the question arises whether OpenAI and/or it's LLM should have signaled that someone's having a bit of a moment that might escalate in the worst way, but that probably happens tens of thousands of times.

Suicide is a common theme. If I remember correctly, "being a suicide survivor" is a theme some people talk about when describing what really triggered the jump start of their career as an artist and whatnot. Ok, but can we say with certainty that that is actually true? Nope. It's just a feeling that draws legitimacy from a post-fucked up mindset.

But that's completely beside the point.

I hate the idea of chat control and polarizing bits like that are just mid-term strategy fuel so I am quite certain that's what it's really about, which it shouldn't be, because some have a looooooooooooooooooooooot of wealth that was mined by the people and should have relieved their lives of a lot of stress but didn't, so, if you want to blame someone: blame Wall Street and Magic Money People, because a) they can take it and b) can do a looooooot about shit like that without jeopardizing the stability of the system. (sidelook)

It's a lawsuit.

OH and BTW, maybe the kid didn't kill himself but there was a MITM who saw a chance to do something reaaally fucked up.

So here's another question: what are the many things ISPs, SWEs, companies and corporations can do to protect customers from MITM without infringing on their rights to privacy?