
I built an LLM modeled after Nietzsche. How accurate is this response?
I’ve been working on an app called The Agora: Philosophy Debates, where I’ve built an LLM modeled after Nietzsche, and the biggest thing I care about is the accuracy and quality of his responses, not just making the model sound like him.
That’s actually why I wanted to post this here. I know a lot of people in this subreddit have spent far more time studying Nietzsche than I have, so I’d really value your perspective.
The model has three different debate modes depending on someone’s experience level. This example is from Balanced mode, which tries to strike a middle ground between Nietzsche’s own style and language that feels natural to a modern reader.
I tested it with this question:
“Is equality a noble or mediocre goal?”
The question is mainly drawn from Nietzsche’s discussions of equality, democratic leveling, hierarchy, and nobility in Beyond Good and Evil.
I’ve attached the model’s response below.
Do you think this is an accurate Nietzschean answer? What does it get right, and what would you change?
Accuracy and quality are my main priorities with this project, so I’m especially interested in anywhere you think the model misunderstands Nietzsche, oversimplifies his position, or puts words in his mouth that he wouldn’t actually endorse.
The app is currently only available on the App Store if anyone wants to test Nietzsche with other questions or try the different debate modes themselves. It’s called The Agora: Philosophy Debates.
Please feel free to leave your thoughts here or DM me. I’d genuinely appreciate any criticism, suggestions, or ideas for what I could change or improve.