Skip to content
IFAI

Industry

Wikipedia bans AI-generated articles

Published Reach 43

English Wikipedia just updated its rules to ban editors from writing or rewriting articles using AI, citing habitual violations of core content policies. Contributors can still lean on LLMs for basic copyediting tweaks and cross-language translations, provided the model doesn't inject original material.

Wikipedia bans AI-generated articles
Image: theverge.com

Wikipedia gave the AI industry the training data to build modern LLMs, and now it is slamming the door on machine-written text.

The free encyclopedia just updated its English-language guidelines to ban contributors from using AI to write or rewrite entire articles, citing the technology's habit of running roughshod over core content rules.

The decision comes after months of volunteer frustration, but the community is stopping just short of a complete blackout.

The two loopholes

Under the updated rules, editors can still turn to large language models for quick copyediting suggestions. The catch is that the model cannot introduce any content of its own into the piece.

Contributors can also use AI to translate articles from other language editions into English. Even then, strict guardrails apply, and editors must be fluent enough in the source language to verify every translated claim before hitting publish.

Interestingly, the guidelines urge editors not to launch witch hunts based solely on vibes. The policy notes that some humans naturally write like LLMs, warning that stylistic or linguistic signs alone are not enough to restrict someone's account.

Volunteers are told to judge work on whether it actually violates core content policies and to review an editor's broader contribution history before taking action.

A months-long cleanup battle

This ban did not come out of nowhere. Wikipedia editors have spent months drowning in automated slop, forcing the platform to set up speedy deletion rules specifically targeting low-effort AI dumps.

Frustrated volunteers even built WikiProject AI Cleanup, a dedicated task force focused on tracking down machine-generated text and teaching others how to spot it.

The formal ban was proposed by editor Chaotic Enby, kicking off a massive community debate that eventually passed with overwhelming support.

The consensus view was simple: clamp down hard on the blatant hallucinations of generative AI, while keeping the door open for harmless, assistive utility.

Why it matters

We see this as a necessary line in the sand for what remains the open web's most trusted reference source.

AI labs rely heavily on Wikipedia to keep their frontier models grounded. If the encyclopedia allowed synthetic articles to flood its pages, the internet would quickly plunge into model collapse, training future bots on the hallucinations of current bots.

By distinguishing between creating facts and polishing prose, Wikipedia struck a very smart balance. It lets contributors polish grammar without outsourcing epistemic authority to statistical prediction engines.

Still, enforcing this policy will be messy. As generative models sound increasingly human, catching sophisticated AI text without wrongly penalizing formal human writers will test the volunteer community to its limits.

Sources (1)

  1. [1]
    theverge.com(opens in a new tab)Originaltheverge.com

Similar events

Past happenings like this one, for context.

Best of AI, daily