Totally scripted conversation.
'Where is Obama's foot?' 'His right foot is on the scale'
Why does the AI only answer for a single foot ?. The fag obama clearly has two feet in that picture. Answer seems completely pre-determined. Or if this is the actual state of AI , it seems quite simple to misdirect. Although to be fair these AI 'engines' run on set parameters with a pre-approved 'vocabularies'. Literally nothing more than a Glorified spell checker. BTW, the Channel that made this video is also hiring in their own pinned comment. Pffat....
I'd ask it many questions about people's feet. Map out what it's grasping. Also ask the same question in different ways. I imagine it would seem pretty shallow after some probing.
Yeah, that's my problem with this. Why is it only vocalizing a determination on a single foot ?? - of course some would say bc that foot is doing something, but that's not actually true - the left foot is just as important if not more for doing something - literally stabilizing the entire body so the right foot can be placed where it is - the argument could be made the left foot is more important in this picture than the right - take the right foot away faggot still stands, take away the left faggot falls down. Question just seems too open ended for the answer and yes you're correct asking more questions to find out what its grasping - but after watching the latest breakthrough AI on a post here last week maybe, parameters of the AI are such that the answer it gives is already kinda pre determined and scripted in its 'available responses' it's not actually grasping anything but running through like three separate programs simultaneously rating 'vocabulary and sentence structure' from a list already programmed into it and splicing it together based on what fits best. Literally a high tech spell checker. Anyway i kinda waffled , don't know if it makes sense how i want it too.
That's the thing, since an AI can study literally billions of responses by humans, even if just a chatbot with canned responses, it can likely find one that sounds very human and relevant. It's like the illusion TV gives, when it's just a bunch of colored dots.
Because it probably assumed he meant the foot that wasn't on the ground because you wouldn't ask that question unless one foot was not in an expected or typical place.
'Because it probably assumed... ' Thats kinda the issue tho, How would it know which foot the person was asking about and why would it assume that without requiring further input/data/ information to clarify before answering. Logically the more important foot is the left one, holding him up from falling - simple biomechanics. Answering 'Where is his foot' is almost an absurdity without requiring more detail. The question is a 101 logical fail. It suggests that the AI was programmed either with (illogical) bias or inputs to essentially to pre-answer, ie scripted. If the AI were to have asked 'Which foot do you mean, obama has two feet in that picture' i would have brought more into this charade.
(post is archived)