Wikipedia volunteers spent years cataloging AI tells. Now there’s a plugin to avoid them. (arstechnica.com)

Tech entrepreneur Siqi Chen has launched an open-source plugin named "Humanizer" for Anthropic’s Claude Code AI assistant, aimed at helping AI-generated content sound more human-like. By utilizing a list of 24 language and formatting cues identified by Wikipedia’s WikiProject AI Cleanup—dedicated to identifying AI-generated articles—this plugin instructs the AI to avoid common pitfalls of robotic writing. Since its release on GitHub, the plugin has gained significant traction, garnering over 1,600 stars in just two days. This development is noteworthy for the AI/ML community as it represents a novel approach to enhancing the quality of AI-generated text by directly addressing the markers that distinguish human writing from AI output. By embedding structured prompts that are more easily interpreted by Claude’s model, the Humanizer plugin optimizes the AI's performance through a clear set of guidelines. Such tools could help increase the trustworthiness and authenticity of AI-generated content, thus bridging the gap between human creativity and machine capability in the ever-evolving landscape of AI language models.