Ai Methodology

Ai was used extensively and unapologetically. It seems almost irresponsible to think about deep, complex ideas without using ai as an interlocutor. That said- the ideas are human, the content is human… and the times where I let two ai models argue two sides of a topic (Impersonal Realism excavation) are explicitly identified and printed as is. Frankly, it was extraordinary to interact, construct, and observe as the models frequently predicted the opposing arguments, narrowed the gaps, sharpened the language, and ultimately converged.

While the platform has a wide range of implications for how we should think about Ai and what it can (or can’t) become, what I want to do here is enable you to use ai to interact with the content. I did print some analysis from each of the 4 leading models which you can read in this section - but no matter how much I insist that I prompted for objectivity- there’s no reason not to suspect I was just manipulating a response… or completely invented it. The solution was to add prompt shortcuts and links so each model can explore the site in an organized way from any angle that interests you. You should see those buttons top right of the screen. The prompt links rely on URLs and the prompt pre-fill which doesn’t always get handled perfectly by the models (especially Gemini for some reason), so I have also created an MCP in Claude that will load the ORC for analysis.