I agree. I for one, welcome LLMs for sensible uses, like trivia, code boilerplate, as a content synthesizer. I wouldn't trust it with anything else, and have f

I agree. I for one, welcome LLMs for sensible uses, like trivia, code boilerplate, as a content synthesizer. I wouldn't trust it with anything else, and have found that any gains obtained by speedups in code writing are offset by the extra cognitive load from digesting code I didn't write.