The Wikipedia of Public Companies

Something strange has happened to financial information. We have more of it than ever — real-time quotes, instant news, AI-generated summaries — yet retail investors understand less than they should about the companies they own.

The problem is not access to data. The problem is access to understanding.

Consider what it takes for a retail investor to truly understand Apple. The 10-K alone is 80 pages. Quarterly filings add another 150 pages per year. Earnings transcripts run 10,000+ words each quarter. A conscientious investor faces hundreds of pages of dense, legally-cautious prose written for institutional analysts and regulators.

So most people don't read any of it. They read summaries instead.

A summary of a 10-K tells you Apple's revenue was $383 billion. It does not tell you why Services revenue matters more than the number suggests, or how the installed base creates a flywheel. Summaries compress; they do not explain.

Consider Apple again. Real understanding is seeing the iPhone as a portal that creates a captive audience, and Services as a tollbooth on that audience. It is understanding that China represents both the largest growth opportunity and the largest geopolitical risk. It is understanding what gross margin trajectory implies about pricing power.

How do you build a system that produces genuine understanding at scale? The answer is not better summarization. It is a fundamentally different architecture — one that separates evidence from synthesis and makes both transparent.

The Vault: Evidence as Foundation

The foundation is not the synthesis — it is the evidence. Before any explanation can be written, a complete repository of primary sources must exist. This is the Vault: SEC filings (public domain), structured financial data, investor presentations, and links to third-party analysis.

The Vault is not just storage. It is the boundary of permissible claims. The synthesis layer can only reference what exists in the Vault. If a fact cannot be sourced to the Vault, it cannot appear in the Wiki. This constraint prevents hallucination by design.

The Wiki: Synthesis with Citations

The Wiki is what readers see: a narrative explanation of the business in plain English. But unlike typical content, every claim carries a citation that traces back to the Vault. The reader can verify anything. Trust is not demanded; it is demonstrated.

There is a secondary vision that extends beyond human readers. As AI agents increasingly participate in financial research — answering questions, conducting due diligence, synthesizing information — they need access to clean, structured, authoritative data about companies.

Today, AI systems crawl disjointed investor relations websites, parse inconsistently formatted PDFs, and guess at which information is current. The result is unreliable outputs that inherit the chaos of their inputs.

A system built on the Vault → Wiki architecture naturally produces something else: a standardized, machine-readable interface to public company information. Clean URLs. Structured data. Semantic content that AI systems can reliably parse. This is not a separate product — it is a byproduct of doing educational synthesis correctly.

When every company has a clean, cited, current explanation of how it works — accessible to both humans and machines — the infrastructure of financial understanding changes. Research becomes more efficient. AI outputs become more reliable. The comprehension gap closes for silicon as well as carbon.

Imagine a world where understanding how Apple makes money is as accessible as understanding how photosynthesis works. Where every publicly traded company has a Wikipedia-quality explanation of its business, updated continuously, with citations to primary sources, free for anyone to read.

In this world, a first-time investor can understand what they're buying — not just the ticker symbol, but the engine. A financial advisor can quickly get up to speed on an unfamiliar company. A journalist can verify claims about a business in minutes. An AI agent can access reliable, structured information about any public company.

This is not a technological fantasy. The primary sources already exist — they're called SEC filings. The synthesis capability exists — it's called large language models constrained by citation requirements. The distribution mechanism exists — it's called the internet. What has been missing is the will to assemble these pieces into a system optimized for understanding rather than engagement, education rather than advice, depth rather than speed.

ValuWiki is one attempt to build toward this future. It is not the only possible approach, and it will not be the last. What matters is the direction: toward a world where the comprehension gap is closed, where understanding compounds rather than decays, where the information asymmetry between institutions and individuals narrows rather than widens.

The comprehension gap is not inevitable. It persists because no one has prioritized closing it — because engagement metrics reward noise, because legal caution produces mush, because the institutions that could build educational synthesis have no incentive to do so.

Closing it requires a different set of priorities: source fidelity over speed, educational depth over engagement, honest dialectic over advocacy, temporal precision over freshness theater, radical accessibility over monetization.

These priorities describe a system that does not yet fully exist. Building it is both a technical challenge and a philosophical commitment — a commitment to the idea that understanding how businesses work should be as accessible as understanding how anything else works.

The Wikipedia of public companies is not a product. It is a standard for what financial education should look like in the age of AI. ValuWiki is one early attempt to meet that standard. The work — of synthesis, of citation, of explanation — continues.