The various examples of questions/answers y’all have noted reminds me that the “answers” are derived from a body of work (religious/philosophical texts) that are yet relatively free from the tainting some of the redditors mention. But how long until some new “Social Gospel According to the WEF” or similar nonsense is pushed into the mix?
This was an interesting thread, so thanks for that, RoosterRedux. AI will never have a soul, so it will never have a true place of standing among men.
I have had some AI platforms argue with me but then relent when I show them how their positions are illogical.
For example, I had a lengthy discussion with ChatGPT about whether moral law is objective and built into reality or is culturally determined and therefore relative.
When ChatGPT took a middle position (that moral law is sometimes relative and sometimes objective) I pointed to the treatment of women under Islam.
At first it tried to defend Islam, describing the unfair and unequal treatment of women as isolated to only a few Islamic cultures. But as we dug into the honor killings in the UK (and elsewhere) and the unfair treatment of women versus men in cases of rape and adultery, it started to weaken.
I got it to eventually apologize.
You mean a Google-programmed AI?
AI will never have a soul, so it will never have a true place of standing among men.
However, since we both have seen AI make errors, and there have been reports of a a few AI models seeking self-preservation, if destruction was programmed if AI ever made an error then would it engage in a HAL 9000 recourse?