Last week, fan fiction aficionados around the world were dismayed to discover that the popular website Archive of Our Own (AO3), which hosts just shy of nineteen million fanfic works, had gone dark. In a statement, the Organization for Transformative Works, the nonprofit which runs AO3, said the site had to be pulled offline due to a “security incident” and that it could remain down for up to a month.
Then AI swooped in and saved the day. And by “save the day” we mean made everything a whole lot worse.
On Sunday, a copycat site going by the domain name AO3.life popped up online. It was very quickly taken down, for reasons we’ll get to in a moment, but while it was live its homepage framed the site as the saving grace of AO3 readers lamenting the loss of their favorite fanfic archive. “Relax. It’s all here,” the homepage says in a screenshot shared on social media. “The Archive of Our Own, saved before it can slip away.” The site was built by a teenage developer in India who used AI agents to scrape original content from A03 so they could sell AI-generated audiobooks based on those materials. Needless to say, many fanfic authors were not psyched about the nonconsensual scraping and monetization of their work.
“Huge violation of trust, consent, and u just fed a BUNCH of authors actual hard work and art to an IP eating, copyright destroying machine,” one Redditor wrote about the copycat site in the AO3 subreddit. “i will never understand what part of this seemed like a good idea. it’s disturbing to see this is how people are interacting with AI and fanfic and somehow believing it is positive.” A number of similar sites, all seemingly based on the mass-scraping of online fanfic archives, have also appeared in recent days, sparking even more consternation from authors.
So severe was the blowback to AO3.life across social media that its creator decided to scrap the whole thing twenty-four hours or so after it had been launched. All that’s left of it now is a hand-wringing apology from that programmer, who identifies themself only as “Ganesh.”
“It was never my intention to hurt so many authors—and I’m truly sorry to you all,” Ganesh wrote in the statement. “There’s no way to show my sorry [sic] better than taking this site down.” Before it was axed, the site told authors that 91% of its revenue would be set aside for them, according to multiple reports. The first three minutes of listening to AI-generated audiobooks were free, but listeners would reportedly be charged to hear more after that. To collect their earnings (or request to have their work removed from the site), authors were asked to submit their AO3-affiliated emails—something that the mods of the AO3 subreddit emphatically warned its members not to do.
In his statement, Ganesh claimed they decided to charge vistors for what they described in their apology as the “AI Slop audiobooks” in order to cover the site’s operating costs. “I doubt anyone knows how expensive it is to run a server like this, especially with the amount of usage this site got,” they wrote, adding it “had terabytes of bandwidth in usage. In an attempt to keep the site up, I thought audiobooks were a good option. I sold like 11 audiobooks, if it’s any consolation.”
Ganesh concluded his apology with what appeared to be heartfelt contrition, and also by setting themself a new goal for the future: keep experimenting, but try to avoid angering legions of strangers. “From a technical standpoint, this was really fun,” they wrote, “I did it all by myself and the time/urgency made me feel really cool. It’s not everyday you have everyone talking about the site you made, but I’ll make sure it’s for something praiseworthy in the future.”