I was concerned about this and have been taking time to show my son how they work. I have ollama on my desktop and, when I have the chance, I'll try my best to dig in and show him peeks behind the curtain.

I'm lucky in that I don't think he has much interest in it, but I still don't want to risk it, so I haven't let him "play with" it.

There's a model that people have been suggesting for a while where kids grow up with their own AI assistant that sticks with them for their entire lives. When people were suggesting this years ago, it was offered in a positive light, but people today would probably have a horror response. I think this is a good thing. It means that we're actually putting some thought into what we're doing here.