Wikipedia bans AI-generated articles

The Wikimedia Foundation has enacted a formal policy prohibiting editors from using large language models (LLMs) like GPT-4 or Claude to write or rewrite articles on the English version of Wikipedia. The update, added to the site's guidelines, states that AI-generated content frequently violates Wikipedia's core content policies, including requirements for verifiability, neutrality, and original research. The policy was proposed by editor Chaotic Enby and passed after a lengthy community discussion with "overwhelming support," targeting what the community sees as "blatantly problematic" uses of AI.

While the ban is strict on authorship, it carves out specific, limited allowances. Editors may use AI tools to "suggest basic copyedits" to their own writing, provided the AI does not introduce new substantive content. They may also use LLMs to assist in translating articles from other language Wikipedias, but only if the editor has sufficient knowledge of the source language to verify the translation's accuracy. The guidelines caution against over-reliance on stylistic analysis to detect AI use, advising instead to focus on an editor's compliance with content policies and their edit history. This formal rule builds on existing community efforts like WikiProject AI Cleanup and policies for the "speedy deletion" of poorly written AI slop.