LLMs Helped Me Understand How I Work

The way Large Language Models (LLMs) work made me reevaluate the way I work. Not “work” in the productivity sense, but “work” in the functional or operational sense.

At the most basic level, an LLM works by guessing what word should come next in a sentence, and it’s making that guess based on what it’s been trained on, which is more or less everything humans have ever made. And if you think about it, that’s almost identical to how you and I would form a sentence. We search our database, weighted against everything we know, and we predict the next word in the sentence.

When I realized this, I was sort of annoyed, or maybe disappointed. Am I just a prediction machine, predicting every next word or moment in my life?

Yes, basically. But, I am afforded some level of credibility because my predictions tend to be less predictable than others. Knowing this, I’ve started to force myself towards less and less predictable outcomes. That’s creativity, and it’s something humans can hope to hold on to for a little while longer.