The problem underlying all of this is the level of trust people are putting into what is essentially a probabilistic engine. An LLM would be more accurately la

The problem underlying all of this is the level of trust people are putting into what is essentially a probabilistic engine. An LLM would be more accurately labeled as a "natural language database" --- not "artificial intelligence". Any apparent "intelligence" is mostly illusionary --- an indicator of the lack thereof on the part of the user.