A copycat fan fiction site that scraped work from Archive of Our Own to sell AI-generated audiobooks went offline about a day after it launched, following sustained anger from the authors whose writing it had taken. The episode shows how quickly a scraping operation can be built, and how quickly writers can force one down.
AO3.life appeared on Sunday, while the real Archive of Our Own was offline. The Organization for Transformative Works, the nonprofit that runs AO3, had pulled the site down over what it called a “security incident” and said it could stay dark for up to a month, Gizmodo reported. The copy site’s homepage framed itself as a rescue for readers missing their favourite archive: “Relax. It’s all here,” it said, according to a screenshot shared on social media. “The Archive of Our Own, saved before it can slip away.”
Gizmodo reported that the site was built by a teenage developer in India who used AI agents to scrape original content from AO3 so that AI-narrated audiobooks could be sold on top of it. Listeners got the first three minutes free and were reportedly charged after that.
Authors push back
Fan fiction writers reacted with anger at the nonconsensual scraping and monetisation of their work. In the AO3 subreddit, one commenter called it a huge violation of trust and consent, and said the project had fed a large number of authors’ hard work to a copyright-destroying machine. The commenter added that they would never understand what part of the idea had seemed good.
Gizmodo reported that a number of similar sites, all apparently built on the mass scraping of online fan fiction archives, have also appeared in recent days, adding to authors’ alarm.
Before it was taken down, the site told authors that 91% of its revenue would be set aside for them, according to multiple reports cited by Gizmodo. To collect earnings, or to ask for work to be removed, authors were told to submit their AO3-affiliated email addresses. The moderators of the AO3 subreddit emphatically warned members not to do so.
All that remains of the site is an apology from its creator, who identifies himself only as “Ganesh.” In it, he wrote that it had never been his intention to hurt so many authors and that he was truly sorry to them all. He said there was no better way to show that sorrow than by taking the site down.
Ganesh described the audiobooks in his apology as “AI Slop audiobooks” and said he had charged for them to cover the site’s running costs. He wrote that he doubted anyone knew how expensive it was to run a server like it, especially given the traffic the site received, and that it had used terabytes of bandwidth. He added that he had sold about eleven audiobooks, in case that was any consolation.
He closed by saying the build had been fun from a technical standpoint, that he had done it all himself, and that the urgency of the moment had made him feel cool. He wrote that it is not every day that everyone is talking about a site you made, and that he would make sure his next project was for something praiseworthy.
The takedown does not settle the wider problem. AO3 itself remained offline as of Gizmodo’s report, and the other scraping sites the outlet described were not said to have been removed. For writers, the practical exposure is the same one the subreddit moderators flagged: any request to hand over an account-linked email address to a site you did not choose to deal with.
Sources
Every quotation and figure in this article is checked against the source reporting linked above before publication. How we report