The reason is a
lack of an immediate verification process. The reason LLMs are been great at programming but less so elsewhere is because if it generates non-working code, calling hallucinated libraries or functions, there's immediate feedback. That code doesn't compile, it crashes, it fails tests, it's slow, it's buggy. But if the LLM gives me bad relationship advice, there's no repl that it can use to retry that conversation with my now-ex until we're back together. Or whatever thing that it's being used for that doesn't have some sort of way to give it a verifier.
So taste is still a thing. Software engineering will have to develop (uh oh) taste and the ability to say no. Your app could include a button to make the device scream like a monkey, but if management tells you it has to have a feature that makes no sense, the answer is no.
So taste is still a thing. Software engineering will have to develop (uh oh) taste and the ability to say no. Your app could include a button to make the device scream like a monkey, but if management tells you it has to have a feature that makes no sense, the answer is no.