AI clearly does not grok what ties are all about...

More about AI clearly does not grok what ties are all about...

I shall admit it, neither do I, really. But I really can't fathom how the video clip it generated has anything at all to do with the action I requested. As in, what on earth is it doing? The knot is still there, and so is the tie, and it seems to be using a ribbon or something to tie up her hair. Has this model never been trained on any video of someone taking off a tie? Is it completely incapable of analyzing the image to see she's wearing one?

Was not a single image and/or video of someone wearing, putting on, or most importantly REMOVING a tie ever used to train this model? Surely some AI developer *somewhere* has to wear a tie now and again, for a meeting or a conference or something, though I suppose it's possible they've suppressed all memory of that. They must have at least SEEN a tie on one of their bosses at some point, or if nothing else the overpaid consultant brought in to tell them everything they were doing (that worked) is wrong, and they must do it THIS way - which doesn't work, and here we are.

Suddenly I have realized. This is just like the Sirius Cybernetics Corporation Nutrimatic Drinks Dispenser. There, you ask for tea, and "it then invariably delivers a cup-full of liquid that is almost, but not quite, entirely unlike tea." I believe I have discovered the root of the problem. AI video generation is another fine product of the Sirius Cybernetics Corporation. Your virtual pal that's fun to be with... this explains so much. Share and enjoy.