Hacker News Reader

Comment by elicash | Hacker News Reader

> simply to make each individual feel good

This seems dismissive. The examples you're replying to relate to "safe spaces" for people discussing fertility, pregnancy, menopause, sexual dysfunction, sexual orientation, gender discrimination, etc. Honestly, some of the dismissiveness I get from this whole thread makes me question whether people lack understanding of some of the complex situations that many people are in.

View on HN

I went back and reread the thread and I see your point. Calling it "feeling good" is certainly oversimplifying. However, I don't think it materially changes my position. Is the woman who is using an AI boyfriend to cope with a verbally abusive RL boyfriend avoiding a problem that she would be better off confronting? Is the woman who has AI children to deal with the grief of not having real children hiding from a grief that will need to be faced eventually? I don't know the details, so I won't venture to say it is so. If it is, though, this is a real problem. We need to come to terms with the brute reality of our existence. The reality is, death and loss are inevitable, other people have their own lives and make choices that hurt us, yet we need them. Perhaps some encouragement and respite through AI can ease this. But it would be very easy for this respite to turn into a refuge, which is one of the mechanism by which drug use becomes drug dependence.

> Is the woman who is using an AI boyfriend to cope with a verbally abusive RL boyfriend avoiding a problem that she would be better off confronting?

Or maybe the woman is getting resources from the chatbot to help her escape the situation.

I don't think we necessarily disagree, since I don't read you as saying there's never any place for AI chatbots. We need independent research that isn't just done by the AI chatbot companies themselves so that folks can figure out optimal approaches for training them.

> I don't think we necessarily disagree, since I don't read you as saying there's never any place for AI chatbots

Definitely not. I can, for instance, see how chatbots might be of great use in palliative care or to help ward off cognitive decline in the elderly.

> We need independent research that isn't just done by the AI chatbot companies themselves so that folks can figure out optimal approaches for training them.

Yes, definitely.

Reply on news.ycombinator.com