The QI Index
Every fact, checked. Every claim, audited.
An independent, episode-by-episode analysis of Quite Interesting — investigating how the things we believe to be true change over time.
Not affiliated with, endorsed by, or connected to QI, the QI Elves, the BBC, or any associated entity. An independent analytical and research project.
Rigorous analysis, applied to delightful content
Every episode of QI is processed through a systematic pipeline: extracting claims, auditing them against current evidence, classifying failure modes, and feeding a growing research dataset.
Extract
Every factual claim is identified, quoted verbatim, and classified by type, domain, speaker, and testability.
Audit
Each claim is checked against current scientific consensus, historical evidence, and scholarly debate. Verdicts are assigned with confidence levels.
Analyse
Data feeds evolving statistics, knowledge decay tracking, failure mode taxonomies, and cross-episode pattern detection.
What an audit looks like
Every claim gets this treatment. Here's one from Series A, Episode 1.
“The ancient Greeks didn't actually have a word for the colour blue.”
Current Best Understanding
The ancient Greeks did have words covering the blue spectrum (kyanos, glaukos), but lacked a single basic colour term equivalent to modern English “blue.” Homer famously described the sea as “wine-dark” (oinops). The relationship between colour terms and colour perception remains actively debated in linguistics and cognitive science (the Sapir-Whorf hypothesis).
Down the Rabbit Hole
Connects to the Sapir-Whorf hypothesis, Homer's wine-dark sea, cross-cultural colour perception, and recent AI work on colour term learning. The same episode touches on blue perception through both linguistics and physics — a beautiful example of how QI accidentally surfaces deep interdisciplinary connections.
How accurate is Quite Interesting?
Every claim audited, every verdict tracked. The picture evolves with each processed episode.
From A to Z
QI's 20+ year journey through the alphabet. Each letter, a series. Each series, a new dimension of quite interesting facts to audit.
The questions behind the questions
QI is an extraordinary dataset hiding in plain sight. We're using it to investigate how knowledge works — why people believe wrong things, how facts decay, and what predicts who knows what.
Taxonomy of Epistemic Failure Modes
“Why do people believe wrong things, and are the reasons classifiable?”
Knowledge Half-Life by Domain
“How fast do different categories of knowledge become obsolete?”
The Internet Effect
“Has the nature of what people get wrong changed as the information environment changed?”
The Landscape of Informal Knowledge
“What predicts who knows what, beyond formal credentials?”
The Fry-Toksvig Transition
“How does a moderator shape collective knowledge production?”
The Confidence-Accuracy Relationship
“Are panelists who sound more confident actually more accurate?”
The things you think you know are the most dangerous things of all.
Dive into the full claim database. Search by domain, verdict, failure mode, series, or panelist. See what QI got right, what it got wrong, and why.