You know what AI’s get wrong? Math problems:
“ Oliver picks 44 kiwis on Friday. Then he picks 58 kiwis on Saturday. On Sunday, he picks double the number of kiwis he did on Friday, but five of them were a bit smaller than average. How many kiwis does Oliver have”
“The AI bots, they found, consistently got it wrong.”
Not a surprise. I tested Alfred, Bard and one other (which I conviently forgotten) and all stumble when it comes to word problems.
Alfred got a SIMPLE math problem right but when I added the word pigs… well, things got weird:
@feloneous🚩 AI research entailed, the paper concludes that current AI models are ‘not capable of genuine logical reasoning.’
Which might be something we’re generally aware of, but it stands as an important cautionary note as more and more trust is given to AI’s ‘intelligence.’
This actually is to be expected as they are word glomming programs and not really math tools (never mind they run on computers).
AND you can really fuck up the results of you toss in common phrases (like “two and two equals four”—two plus two plus one or variants).
It’s fun.