Fan tribute vs plagiarism: How an AI-driven book website tricks search engines, chatbots, and even loyal fans

A San Francisco marketing agency copied the entire contents of a bestselling book, replaced its original artwork with AI-generated images, and launched a website that now outranks the official page in search results. Andy Baio writes on Waxy.org that the agency, Qontour, built the unauthorized platform for John Koenig’s “The Dictionary of Obscure Sorrows” without the author’s knowledge or consent.

Koenig, who spent over a decade creating the collection of made-up words for undefined emotions, told Baio: “Yeah man, I had nothing to do with it. Don’t know what to think or do about that, as the site is pretty slick. Nicer than my own, really.”

The website includes the full text of the book, all 311 neologisms with definitions and essays, plus AI-generated illustrations made with DALL-E 2 that contain typical visual errors. A “Submit A Sorrow” feature invites visitors to describe a feeling and uses OpenAI’s GPT-4 to auto-generate a new word, its etymology, and a definition. These user submissions land in a public gallery, licensed under CC Zero.

Qontour, previously known as Prompt Digital, calls the project a fan tribute but embedded its own Amazon affiliate code throughout the site, earning commissions on book sales. The agency also prominently features the site in its portfolio, advertising its expertise in “AI-powered image libraries” and “extensive content integration.”

Copyright Confusion

Simon & Schuster, the book’s publisher, filed two DMCA takedown requests with Google last year. Neither action reduced the bootleg site’s visibility. Today, that site appears as the top result for searches of the book title, Koenig’s name, and individual coined words. AI chatbots such as ChatGPT and Gemini also reference it as the official source, falsely attributing the entire project to Koenig and the agency together.

This incident exemplifies a growing pattern: individuals and companies use generative AI to repackage copyrighted human work, often displacing the original creator in search rankings and AI answers. For content professionals, it underscores the need for clear consent frameworks and proactive brand protection when AI can easily scrape, remix, and redistribute their intellectual property.