AI Has a Huge Problem (And a Man From 1920 Explains Why)
AI Has a Huge Problem LLMs have “infinite context”—trained on the entire internet. They’ve seen everything. Medical journals. Legal documents. Reddit threads. Stack Overflow. Poetry. Garbage. And they hallucinate. Miss the point. Treat everything with equal weight. They remember everything but understand nothing. When you ask an LLM to “generate content,” it pulls from everywhere. No filter. No priority. No sense of what matters for YOUR specific situation. This is the fundamental problem with prompts. ...