As artificial intelligence increasingly influences the editorial and media sectors, websites are creating rules to govern its use. This week, Wikipedia announced that its editors are prohibited from using AI-generated text to create or rewrite articles, though the platform did not entirely ban AI from editorial processes.
In its recent policy update, Wikipedia clarified that “the use of LLMs to generate or rewrite article content is prohibited,” refining previous, less precise guidance that stated LLMs “should not be used to generate new Wikipedia articles from scratch.”
The use of AI in Wikipedia articles has sparked debate within the site’s large volunteer editor community. According to 404 Media, the updated policy approved through an editor vote passed overwhelmingly, with 40 votes in favor and 2 against.
The guidelines do allow limited AI use for specific editorial tasks. Editors may use large language models (LLMs) to propose basic copy edits to their own writing.
These suggestions can be applied after careful human review, provided the LLM does not add new content.
The policy emphasizes caution, noting that LLMs can sometimes exceed instructions and change the meaning of text, potentially causing discrepancies with cited sources.
