The Archive of Our Own (AO3) clone site ao3.life is offline after its creator received more than 2,000 takedown notices in a single day.
The website, launched last week during the platform’s outage, scraped fanfiction works and offered them as EPUBs or paid AI audiobooks. It went down on Monday evening after widespread backlash from readers and authors on social media.
The site’s creator, Ganesh, explained that the project used a swarm of AI agents controlling 40 virtual machines to scrape the internet for content. The homepage had promised to save works before they disappeared, allowing users to read them or let the site read aloud when their eyes grew tired.
AO3 went down on Thursday following a security incident. The Organization for Transformative Works (OTW), which owns the repository, expects the site to remain inaccessible for up to four weeks. OTW states no data has been lost, but the platform hosts nearly 18.5 million works. Authors and readers are currently unable to access the site.
ao3.life offered works as EPUBs for e-readers and as AI-generated audiobooks. Users could listen for free for the first three minutes before being charged. The site claimed authors would receive 91% of profits, with 9% going to cover narration and processing costs. It stated that earnings would wait indefinitely until the author collected them, even without an account.
On AO3, authors can archive lock their works to make them visible only to logged-in users. This prevents AI bots from scraping them and gives authors control over how far a work spreads. However, authors could not access AO3 while it was down to lock their works if they were publicly available before the outage.
Some authors reported that archive locking did not stop the scraping. One writer posted on Bluesky that they had locked everything on AO3, yet their works were still present on the clone site with zero control over them.
The clone site stated authors could request removal by emailing le***@*o3.life. Proving authorship required leaving a note on an AO3 profile for confirmation. This was problematic because AO3 remained completely down for the duration.
Many authors clearly used the legal complaints email. Ganesh wrote that they facilitated over 2,000 takedown requests in the time the site was up. They stated it was never their intention to hurt authors and that there was no better way to show their apology than taking the site down.
Ganesh questioned why they charged for AI audiobooks, noting the high cost of running a server with the amount of usage the site received. The site had 126,592 unique users within 24 hours of launching and consumed terabytes of bandwidth. They sold only 11 audiobooks, which they offered as consolation.
Fanfiction has become a target of AI scraping and repackaging in recent years. In 2024, I reported on how Etsy sellers turn free fanfiction into printed books and list them for sale for as much as $100 a copy.
With AO3 down, some people have started sharing Google Drive links on Reddit and Discord containing works downloaded from the site previously. This has alarmed and angered many writers. Whether authors have the right to profit from their works depends on the content and their use of existing intellectual property. Fanfiction is typically considered fair use, but selling it could draw the ire of lawyers and a potential lawsuit even if the author ultimately wins.
Authors on AO3 are prohibited from monetizing their works or linking to payment platforms. Ganesh stated it was never their intention to hurt authors and hoped subsequent actions represented that. They signed the note as a 17-year-old from India who thought they were doing something good.
What it means
This incident highlights the legal and ethical risks of scraping copyrighted or licensed content without permission. The failure of the site demonstrates the difficulty of monetizing fanfiction through AI tools when the original platform is unavailable and authors actively object.
