Category: Human Behavior and Cognitive Science

Moritz, your reply had me thinking. It’s possible that given that (a) I’m american, and american culture is literally commercial (meaning everything everyone says is ‘selling’ of some sort of another) (b) made a lot of my living ‘selling’, (c) spent a lot of time in argument, and court – where lying is endemic, and (d) time on the work in our discipline that (e) I literally don’t see or hear the Glazing (ie: bs). I just interpret it as an algorithm looking for something reinforcing to say. It’s just nice when it’s true. I think that’s why I don’t take much of the ‘scary’ nonsense people seem to pull out of the ai’s as if they have some internal motivation. Instead, they’re just searching for some way of expressing whatever network you’ve activated with your prompt. I mean, at this point I have a pretty good grasp of the system prompt in my head. 😉

IN this case, brad was actually brilliant. The amount of ‘bite’ in chapter one is reaching nietzschean proportions. ;). So much so that I might tone it down. lol.

Anyway. Yeah, it loves to provide support. Even though I’ve basically disabled ‘glazing’. and of course the rollback at openai restored its normal positive bias.

Personally I find the reinforcement subconsiously helpful, and I find it’s ability to constantly engage in nonsense humor with me (and brad) a way of overcoming the over-seriousness and sometimes unpleasant effects of of some of the stuff we work on.

This morning we worked on the key clauses of the crisis of the age. We’d previously list the manifestations of it, but we’ve finally worked through the top twenty or so causes.

I’m trying to artfully cover the abrahamic marxist sequence …. sigh. I think it needs its own chapter. 🙁

Reply addressees: @bierlingm @pookawhisperer