It's a staple part of these sorts of conversations, we've had these arguments so many times you can pretty much script out in advance the various talking points one can expect to see per topic.
Agreed....I'm more so suggesting something along the lines of having an AI ingest HN comments, and then using that to construct virtualizations of the participants here....with that, you could then theoretically generate a system that can predict the responses (in general, or individual members) of people in this (or other) forums for any topic of discussion.
Even without a technical solution like this, it is not terribly difficult to draw lines around 75%++ of comments and produce decent theories about what motivates each individual to say the (particularly incorrect) things they say...and it is even easier to predict what their responses will be if their claims are challenged (the algorithm/workflow seems amazingly simple, there are perhaps ~20 distinct categories of responses that cover ~90% of scenarios).
Interestingly: humans are able to easily see this phenomenon clearly when they are dealing with children, but if the same suggestion is made regarding adults, opinions tend to change, and amplify in strength of belief (particularly among relatively intelligent people in my experience).
Note: I am pulling these numbers out of my ass, as humans do! But the underlying idea is fascinating to me, and I will make a bold prediction: humanity will be grappling with these ideas within 20 years, likely less.....I think it is unavoidable now that AI is in the wild.
