I spent my 42nd birthday generating 100% pure AI slop.

I adapted Machiavelli The Prince to the Silicon Valley context, using Sonnet. It took a 6-hour+ non-strop spree to write a 81 pages, and it's kind of an instructive read!

Rather than making any line edits, I was able to share high level feedback over the first few chapters and keep regenerating. This process was awesome - I could be extremely ruthlessly critical for hours on end and it didn't hurt Claude's feelings, they just improved, until it was effectively "trained". We're all used to that with objective contexs like coding, but in a subjective context it was a different and really good feeling, like working with an incredibly mature friend in your band.

Because philosophical texts are short, you can write adaptations that maintain your feedback to the end, making it easier and easier to generate later chapters with no intervention.

One of the better collaborative experiences I've had! Once we have agents ready, should kind of just adapt all classic philosophical texts to allegories about DHH, AMD, and uh, Quibi.

Enjoy!