Spyglass is a historical atlas for the AI age — an interactive tool for exploring human civilization from the dawn of agriculture to 2020, built on a rich custom dataset and a map-based UI with an AI-powered guide.
Explore the Atlas →You're reading about the Roman Republic in 200 BCE and a thought crosses your mind: what was happening everywhere else? In Han China, in Maurya India, in Mesoamerica. Traditional references bury that context across hundreds of separate articles. You can get there eventually, but you're clicking through six degrees of Kevin Bacon just to build a picture that should be immediate.
Spyglass started with a simple product insight: there is a better way to explore history than walls of text. The depth of a reference work, the immediacy of a visual interface, and the flexibility of a conversation — all in one place. The result is richer exploration than text alone, with chat-based insight that makes every connection one degree away when you want it.
Spyglass gives you a map of the world at any point in the last twelve thousand years — from the first permanent settlements around 10,000 BCE through 2020. Cities appear and grow, empires rise and fall, and you can click into any of it.
Every civilization is profiled across six dimensions: governance, economy, society, culture, science, and military. Each is scored on structured rubrics that define what a 1 versus a 5 means in historical context — so that comparisons across time and place are meaningful, not arbitrary. Every score is grounded in narrative arcs that tell the human story behind the numbers.
On top of the visual atlas sits an AI-powered chat that doesn't just answer questions — it navigates the atlas for you. Ask "Show me the Song Dynasty at its peak" and it jumps to the year, selects the city, and opens the profile. Ask "Compare the Aztecs to the Inca" and it pulls both. The chat reads from the live dataset — it doesn't guess, it queries.
Nothing like this dataset existed off the shelf. Over 2,000 entity-period records, each with 64 custom attributes — governance type, bureaucratic sophistication, trade networks, financial systems, literacy, gender autonomy, scientific paradigm, military capability, and much more. 700+ places with population timelines stretching back millennia. 11,000+ population snapshots with confidence ratings so you always know how much to trust a number. 12 world regions, with particular care given to parts of the world that English-language references tend to underserve.
It's the kind of structured, scored data that makes it possible to place the Abbasid Caliphate next to Tang Dynasty China and actually learn something. You can explore the full scoring methodology on the rubrics page.
Spyglass uses AI to research, score, and generate narrative content for its dataset. The process is structured — scores are assigned against published rubrics, narratives are grounded in quantitative data, and validation catches inconsistencies — but AI can make mistakes.
Historical population figures, governance scores, and narrative arcs should be treated as informed estimates, not definitive scholarly claims. Population data before 1500 CE carries particular uncertainty — confidence ratings (low / medium / high) indicate how much weight to give each figure. Scores represent structured approximations anchored to rubric definitions, not precise measurements. We welcome corrections — if something looks wrong, let us know.