Honestly, that’s the worst way to use ChatGPT. It works best with content you’ve already written or with well-developed prompts with very specific parameters.

The problem isn’t the technology ... it’s how people are using the technology.

I also don't think I would trust what it says about it's most used words. It doesn't have self-awareness, nor does it statistically analyze its own output (that I know of).

That said, I get what you are saying about frequent tropes. ChatGPT loves the metaphorical journey ... but that's because we overuse that metaphor in its dataset.

It's not stealing anything ... it's reflecting ourselves back to us.

Expand full comment