I asked three different commercially available LLMs the same question: Which TLDs have the same name as valid HTML5 elements? This is a pretty simple question to answer. Take two lists and compare them. I know this question is possible to answer because I went through the lists two years ago. Answering the question was a little tedious and subject to my tired human eyes making no mistakes. So…
Frankly you’re just holding it wrong if you expect it to do actual data processing like this directly.
Ask it to write a Python script to compute the result and it will get it 100% right.
A spoon being a poor shovel would not be such a big issue, if it wasn’t advertised as a great shovel that will revolutionize the construction industry, and if all the construction companies weren’t stockpiling spoons and telling their employees they will be fired if they don’t use the spoon
I don’t think I’ve seen anyone really promote AI as a solution to something that could trivially be better done by ordinary programs. Nobody is saying we should replace simple maths with AI. There’s no AI sorting algorithm. No AI DNS server. No AI ALU.
AI is being pushed in places where we don’t have any other automated solution - writing code, natural human interaction, reading and writing prose, etc.
AI vending machines…
Those seem to be like Amazon Go, where you just take what you want and it magically charges you.
Whether that is worth avoiding pressing two buttons or not is debatable (totally isn’t), but it is definitely something you can’t easily do without AI.