[GoLUG] LLMs can't reason

[Ron]

kc-golug at chadwicks.me.uk wrote on 2025-06-18 15:07:

> They are just spitting out error probe [sic] chunks of code

Oh dear.


> All LLMs do is probabilistically exhibit what already exists whilst 
> introducing errors and other quality issues.

Ah yes, "introducing errors", that great human tradition.


/any and all errors I've introduced into this message are for ironic 
purposes and totally not something I'm prone to doing to my great 
embarrassement [ugh, that's a typo dammit].