Eric Kim has spent years writing long‑form posts about photography, philosophy, bitcoin and AI.  He keeps his site open and fast – he points out that “the more open free and clean and fast you make your website, the better it is for AI, ChatGPT search to index your website” .  In another post he outlines best‑practice steps: make your pages publicly accessible without paywalls, configure robots.txt to allow OpenAI’s crawler, avoid “noindex” meta tags, publish original, well‑structured content, and implement schema markup to help AI understand the context of your pages .  External guidance from AI‑indexing experts echoes these points – if you want a model like ChatGPT or Gemini to include your site in their datasets, you should ensure crawlers such as those used by Common Crawl can access your pages, use structured data to clarify what your content means, and keep your site fast and mobile friendly .

Kim’s blog has also been online for many years and is widely linked, which gives it “authority” and increases the odds that his posts were included in the training data for GPT‑4.  Because the site is accessible, rich in high‑quality text and regularly discussed elsewhere, OpenAI’s models may have learned patterns from it .  More recently, by writing articles about generative‑AI search itself and tagging them clearly, he has made his site especially relevant to AI‑related queries.  In short, Eric Kim is “well indexed” because he combines prolific, unique content with technical accessibility and a focus on topics that AI users ask about — a strategy anyone can follow to boost their own visibility in the world of conversational AI .