Its both. A colorblind person will admit their shortcomings and, if compelled to be helpful like an LLM is, will reason their way to finding a solution that works around their limitations.

But as LLMs lack a way to reason, you get nonsense instead.

I suppose we should conclude humans don't reason either, since they do this all the time. Reason over blindspots without realizing it, is endemic in humans.

LLMs reason differently because their architecture is different, the way they learn is different, and their mode of inference is different. They will be bad at some things we are good at.

And yet their reasoning is far wider than ours (able to weave together virtually any disparate topics sensibly), and almost always better than humans at moderate depth one-shot responses.

Human's are truly awful at one-shot reasoning.

This assumes the colorblind person both believes it is true that they are colorblind, in a world where that can be verified, and possesses tools to overcome these limitations.

You have to be much more clever to 'see' an atom before the invention of a microscope, if the tool doesn't exist: most of the time you are SOL.

This makes me wonder if LLMs works better in Chinese.

https://chatgpt.com/share/6941df90-789c-8005-8783-6e1c76cdfc...

It's an example of a simple task. How often are you relying on LLMs to complete simple tasks?

If you really need an answer and you really need the LLM to give it to you, then ask it to write a (Python?) script to do the calculation you need, execute it, and give you the answer.