This Wikipedia advice page is a field guide to patterns that may indicate undisclosed AI-generated writing in articles, drafts, comments, and edit summaries. It groups examples across content, language, formatting, markup, citations, and communication. Common patterns include generic praise of a subject’s importance or legacy, vague claims of association, promotional phrasing, formulaic conclusions, repeated “not just X, but Y” constructions, dense use of fashionable vocabulary, excessive bolding, awkward headings, broken citations, and canned assurances in edit summaries. The guide also notes possible signs of human authorship, ineffective indicators, and clues that have become outdated as AI tools and their habits change.
The central caution is that these patterns are observations, not rules: none alone proves text was generated by AI, and the page is advice rather than Wikipedia policy. Human writing can share the same traits, while AI detectors and people both make substantial identification errors. The guide warns against simply editing away visible stylistic clues, since they may point to more consequential problems such as vague or exaggerated claims, unreliable sourcing, or policy violations. Editors should investigate and address those underlying issues, or flag them for review; only a narrower set of clear indicators meets Wikipedia’s speedy-deletion standard for unambiguously AI-generated pages.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.