SCHH Community Guide

How Deep Dives are made

We're upfront about it: AI does the research and writing here, through a four-phase process built for accuracy rather than speed.

  1. Independent research. Three leading AI models — currently Anthropic's Claude Fable 5, OpenAI's GPT‑5.6 Sol, and Google's Gemini 3.1 — each perform deep research on the same topic, independently, from an identical prompt. None sees the others' work.
  2. Synthesis and verification. The three reports are compared side by side: points of agreement are identified, discrepancies are resolved or flagged, and important claims are verified against the sources they cite. The result is a consolidated research package.
  3. Writing. A separate model turns that package into an accurate, engaging narrative for a general community audience — without introducing any fact that isn't supported by the package.
  4. Fact-check. Another model then checks the finished article against the research package and the cited sources, correcting errors, overstatements, and qualifications that got lost in the writing.

Even so, no process is perfect — if you spot something wrong, tell us and we'll fix it and note the correction.

Suggest a topic

What should we dig into next? Anything about the community and the place we live is fair game — how something here came to be, how it really works, what it costs, what's changing. Send your idea through the contact form and it goes straight on the research list.