A Wikipedia Group Made a Information to Detect AI Writing. Now a Plug-In Makes use of It to ‘Humanize’ Chatbots

On Saturday, tech entrepreneur Siqi Chen launched an open supply plug-in for Anthropic’s Claude Code AI assistant that instructs the AI mannequin to cease writing like an AI mannequin.

Referred to as Humanizer, the easy immediate plug-in feeds Claude a listing of 24 language and formatting patterns that Wikipedia editors have listed as chatbot giveaways. Chen revealed the plug-in on GitHub, the place it has picked up greater than 1,600 stars as of Monday.

“It’s actually helpful that Wikipedia went and collated an in depth checklist of ‘indicators of AI writing,’” Chen wrote on X. “A lot with the intention to simply inform your LLM to … not do this.”

The supply materials is a information from WikiProject AI Cleanup, a gaggle of Wikipedia editors who’ve been looking AI-generated articles since late 2023. French Wikipedia editor Ilyas Lebleu based the challenge. The volunteers have tagged over 500 articles for evaluate and, in August 2025, revealed a proper checklist of the patterns they saved seeing.

Chen’s software is a “ability file” for Claude Code, Anthropic’s terminal-based coding assistant, which includes a Markdown-formatted file that provides a listing of written directions (you may see them right here) appended to the immediate fed into the big language mannequin that powers the assistant. Not like a traditional system immediate, for instance, the ability data is formatted in a standardized approach that Claude fashions are fine-tuned to interpret with extra precision than a plain system immediate. (Customized expertise require a paid Claude subscription with code execution turned on.)

However as with all AI prompts, language fashions don’t all the time completely observe ability information, so does the Humanizer really work? In our restricted testing, Chen’s ability file made the AI agent’s output sound much less exact and extra informal, nevertheless it might have some drawbacks: It received’t enhance factuality and would possibly hurt coding capability.

Specifically, a few of Humanizer’s directions would possibly lead you astray, relying on the duty. For instance, the Humanizer ability contains this line: “Have opinions. Don’t simply report details—react to them. ‘I genuinely don’t know methods to really feel about this’ is extra human than neutrally itemizing execs and cons.” Whereas being imperfect appears human, this sort of recommendation would in all probability not do you any favors for those who have been utilizing Claude to write down technical documentation.

Even with its drawbacks, it’s ironic that one of many net’s most referenced rule units for detecting AI-assisted writing could assist some individuals subvert it.

Recognizing the Patterns

So what does AI writing seem like? The Wikipedia information is particular with many examples, however we’ll offer you only one right here for brevity’s sake.

Some chatbots like to pump up their topics with phrases like “marking a pivotal second” or “stands as a testomony to,” based on the information. They write like tourism brochures, calling views “breathtaking” and describing cities as “nestled inside” scenic areas. They tack “-ing” phrases onto the tip of sentences to sound analytical: “symbolizing the area’s dedication to innovation.”

To work round these guidelines, the Humanizer ability tells Claude to switch inflated language with plain details and gives this instance transformation:

Earlier than: “The Statistical Institute of Catalonia was formally established in 1989, marking a pivotal second within the evolution of regional statistics in Spain.”

After: “The Statistical Institute of Catalonia was established in 1989 to gather and publish regional statistics.”

Claude will learn that and do its finest as a pattern-matching machine to create an output that matches the context of the dialog or job at hand.

Why AI Writing Detection Fails

Even with such a assured algorithm crafted by Wikipedia editors, we’ve beforehand written about why AI writing detectors don’t work reliably: There’s nothing inherently distinctive about human writing that reliably differentiates it from LLM writing.

One motive is that though most AI language fashions have a tendency towards sure forms of language, they may also be prompted to keep away from them, as with the Humanizer ability. (Though generally it’s very tough, as OpenAI present in its yearslong battle towards the em sprint.)