A look at Wikidata, a sister project of Wikipedia, whose bot-friendly database supports many AI assistants like Alexa
HUMANS PRICKED BY info-hunger pangs used to hunt and peck for scraps of trivia on the savanna of the internet. Now we sit in screen-glow-flooded caves and grunt, “Alexa!” Tweets: @wikimediauk , @wired , and @lilianedwards Tweets: Wikimedia UK / @wikimediauk : Wired calls @Wikidata ‘an obscure sister project of Wikipedia’, but while that may be relatively true, the big upsurge in coverage of it in the past year tells us that Wikidata is becoming increasingly important. http://www.wired.com/... @wired : Wikidata, an obscure sister project to Wikipedia, aims to (eventually) represent everything in the universe in a way computers can understand. http://www.wired.com/... Lilian Edwards / @lilianedwards : Inside the Alexa-Friendly World of Wikidata ; this sounds rather eerily like Semantic Web Project - are they connected? http://www.wired.com/...
Context & Ripple Effects
This 2019 Wired profile caught Wikidata at an inflection point: a project even sympathetic observers like Wikimedia UK called 'obscure' was suddenly newsworthy because its bot-friendly, structured database had become a fact source for AI assistants like Alexa. The piece frames the consumer side of that shift — trivia-seeking moving from search-and-click to voice queries answered from structured data.
The arc since then validates the profile's premise and exposes its cost. Wikimedia Deutschland has kept building for machines, launching the Wikidata Embedding Project with nearly 120M entries for semantic AI access, while the human front door is eroding: Wikimedia now reports significant declines in Wikipedia's own traffic as people get answers via chatbots instead of clicking through.
First-order effects
- Alexa and similar assistants draw directly on Wikidata's volunteer-maintained structured facts, meaning Wikimedia's editors are already doing unpaid infrastructure work for commercial AI products.
- For users, factual lookup shifts from browsing Wikipedia pages to asking assistants whose answers are assembled from Wikidata records — no page view, no attribution surface.
Second-order effects
- Wikimedia Deutschland responded to machine demand by building dedicated AI-access layers like the Embedding Project, effectively running a two-track operation: one interface for humans, another for models.
- That same machine channel undercuts the flagship: as chatbots answer from Wikimedia data, Wikipedia's declining human traffic weakens the reader attention and community pipeline the whole project depends on.
Third-order effects
- If the pattern holds, Wikimedia becomes critical AI commons infrastructure governed by an editor community increasingly at odds with machine consumption — visible in the English Wikipedia's ban on AI-written articles and the Foundation pausing its AI-generated summary experiment after editor backlash.
- The structural question the profile raised in 2019 — who funds and governs knowledge when machines, not humans, are the audience — is now the central tension in how AI sources the world's facts.
The trend: Volunteer-built knowledge commons like Wikidata are being repurposed as machine-readable substrate for AI assistants, decoupling the value of curated facts from the human traffic that sustains them.