Asger Risom
885,000 AI-generated pages, chat-to-edit promised, with no human oversight or citations.
885,000 AI-generated pages, chat-to-edit promised, with no human oversight or citations.
What it is
Elon Musk’s xAI has launched Grokipedia 0.1: an encyclopedia written and “reviewed” by the Grok chatbot itself.
Scale without safeguards
The site debuts with 885,000 articles — but no independent editors, source lists, or fact-checking workflow.
Built by the same model that judges itself
Grok generates and verifies entries, a closed loop that invites AI hallucinations to pass as “truth.”
Conversational editing coming
Users will soon ask Grok to add, change, or remove content — edits mediated by the same model, not a community.
Borrowed backbone
Many entries adapt text from Wikipedia under Creative Commons, undercutting the “from scratch” narrative.
Wikipedia’s answer
Wikimedia’s response was dry but pointed: alternatives come and go — and even Grokipedia depends on Wikipedia.
Openness with asterisks
Code and data are “open,” but moderation rules, audit logs, and provenance of claims are unclear or absent.
Bias and control
With one company’s AI deciding what’s included and how it’s worded, systemic bias and agenda-setting are baked in.
Accountability gap
If a page defames, misleads, or causes harm, who’s responsible — the model, the platform, or no one at all?
Not a drop-in replacement
An encyclopedia is more than text at scale; it’s sources, debate, and transparent revision history — none are evident here.
What to watch next
Will Grokipedia add citations, human oversight, and verifiability, or double down on automation over accuracy?
This article is made and published by Asger Risom, who may have used AI in the preparation
