(May 14, 2023 at 2:02 am)Rev. Rye Wrote: I can just imagine ChatGPT being asked about how to be like Lester Ballard from Cormac McCarthy's Child of God.
And for those of you who haven't read the book or seen the movie, well...
That's good! I don't know that movie at all, but when I was trying to think of characters with -- shall we say -- problematic moral messages, I did think of the villain from No Country for Old Men. I don't know the books well enough to figure out if there's a subtext, or if it's just a portrayal of violence. I do think that a lot of modern people in the US will feel admiration for the characters' successful use of violence.
So it puts the AI in a difficult spot. I suppose what it comes down to is legal liability. If the AI were to recommend or praise some behavior that got someone into trouble, a lawsuit would be inevitable. So they have to keep their analysis as anodyne and school-teacherish as possible.