Originally posted by Lanny
Oh absurdity-as-attention-grab is definitely something actively explored by ML based advertising. I mean I think humans were doing this before machines got into it, but definitely more interesting when you can try out a thousand variations a second to see what sticks. Kind of a natural progression from the “obviously wrong rage-bait” genre.
I think in the next 20 years we’re tragically going to learn more about the human brain from data mining and successful advertising strategies/funnel analysis than like real neuroscience
I actually don't think it has anything to do with absurdity really. It is hard to say exactly but it is like they will hit all the buzzers on what might grab ones attention or engagement while completely bypassing the "making semantic sense" stage entirely.
You could say it is absurd but there feels like some even colder and tighter logic that AI ads have managed to decode.
It is like listening to Oingo Boingo song where it initially seems like a whole song's worth of ideas but then you realize Danny Elfman actually just made some cool 5 second riff that he circled around for 3 minutes. When you identify that riff, then suddenly it is like you understood the master key to the whole song and the logic to the whole thing is actually conceptually even more simple than you initially expected.
I wrote part of a short story on here about a "convincing machine". I find that idea simultaneously terrifying and beautiful. We are, as creatures, essentially just open to being "hacked" because of language. If you find the right combination of sigils to put on my screen, one can be made to believe anything.
But why on Earth should the only paths to making me behave a certain way, have to take routes that make semantic sense to me? There is probably a way to "inception" anything into anyone's mind and it wouldn't need to make sense at all to them.