An anonymous X account named @heatedrivalryai posted a skin for Archive of Our Own (AO3) on June 29 that purportedly identifies coding artifacts left by Anthropic's Claude bot. The development follows a movement in fanworks communities aimed at identifying authors using generative AI.

According to the @heatedrivalryai account, "When a Claude-generated response is pasted directly into AO3 from Claude, the text is wrapped by a Claude-injected code 'font-claude-response-body.'" The account stated, "Its presence indicates the use of Claude definitively." When a user visits a page with the specific code, the skin turns the entire background red.

Several test posts were published to Archive of Our Own to allow users to check if the skin works. The code wrapping is only preserved if text is copied directly from Claude into Archive of Our Own's editor. The tool will not detect text that was edited in Google Docs or Microsoft Word and then moved to Archive of Our Own. Some writers who have been flagged by the tool have updated their works to remove the artifacts.

Anthropic did not respond to a request to verify if the fan-made Claude detector works as described. Google and OpenAI have not responded to requests regarding whether their models leave traceable artifacts in text generation. Archive of Our Own has a tagging system that includes a "Created Using Generative AI" tag, which many authors use to disclose the use of tools like Claude.

Separately, at least one person claims to have written separate code that can detect usage of Claude, Deepseek, and some ChatGPT models. The person claiming to have written code for multiple AI models has not released the solution to the public or explained how it works.