If humans can find bugs, why can't humans write flawless code? Whatever the answer to that conundrum might be, LLMs are trained on these patterns and replicate

If humans can find bugs, why can't humans write flawless code? Whatever the answer to that conundrum might be, LLMs are trained on these patterns and replicate them pretty faithfully.