Authors turn to writing log tools as AI accusations spread

In short: More writers are using tools that record their writing process to defend themselves when people accuse them of using AI.

What's going on

Accusations that books are written by AI are becoming a recurring problem in publishing. In one widely discussed case, Hachette Book Group canceled a planned US release of the horror novel Shy Girl and discontinued it in the UK after online speculation and reporting raised doubts about how it was written. The author, Mia Ballard, denied using AI and said an editor made AI-assisted changes to an earlier version.

This kind of controversy can move fast, and it can hurt an author even if the claim is wrong. Author Andrea Bartz, who previously sued AI company Anthropic over alleged use of pirated books for training, warned that AI fears can turn into a “witch hunt.” She says there is no shared, trusted way to prove a book was written by a human.

In response, several software products are pitching themselves as a kind of receipt system for writing. Services such as Workings, Blindproof, Wintertext, Human Ink, and ObraVera say they can track drafts and time stamps, or add unique codes to files (like a tamper-evident seal on a document). The goal is to give authors a “paper trail” they can point to if readers, agents, or publishers question them.

What to watch

Both AI detectors and “paper trail” tools have limits. Detection tools can flag human writing by mistake, and writing logs can be faked if someone retypes AI text. The Authors Guild says it does not endorse one tool, and it argues writers should not be forced to keep receipts just to be believed.

Source: Wired