> How is a chatbot supposed to determine when a user fools even themselves about what they have experienced?
replyAnd even if it _could_, note, from the article:
> Overall, the participants deemed sycophantic responses more trustworthy and indicated they were more likely to return to the sycophant AI for similar questions, the researchers found.
The vendors have a perverse incentive here; even if they _could_ fix it, they'd lose money by doing so.
> Maybe it's not so sensible to offload the responsibility of clear thinking to AI companies?
replyMarkets don't optimize for what is sensible, they optimize for what is profitable.
It's not market driven. AI is ludicrously unprofitable for nearly all involved.
replyThe profit appears to be capturing the political class and it's associated lobbies and monied interests.
reply> clear thinking
replyMost humans working in tech lack this particular attribute, let alone tools driven by token-similarity (and not actual 'thinking').
deleted
replyIt's almost as if being a therapist is an actual job that takes years of training and experience!
replyAI may one day rewrite Windows but it will never be counselor Troi.
Implying that programming is not an actual job that takes years of training and experience
replyTo be clear I don't think the AI can do either job
Well, unless insurance companies figure out they can make more money by pushing everyone onto AI [step-]therapy instead of actual therapy
replyCome on, I'm sure Dario can find a nice tight bodysuit for claude
reply