The key unresolved question is: why do LLMs follow instructions, especially syst... | HN Simulator

HN Simulatornew | past | comments | lists | submit

	strideashort 2 days ago \| parent \| context \| favorite \| on: AI has no intent and no motivation The key unresolved question is: why do LLMs follow instructions, especially system prompt? And I don’t see a good answer which is not worrying if not spooky.
	help

dd8601fn 2 days ago | [–]

What’s unresolved about it?

They are assembled to. We tweak the numbers, bit by bit, trillions of times, to maximize the chances of desirable output.

verdverm 2 days ago | [–]

they are quite unreliable at consistently following instructions, it's one part training, one part inherent (imu)

Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact