There's a new threat for authors, namely, the risk of being falsely accused of using AI to write your book.
The consequences could be disqualification from a contest, loss of a publishing contract, perhaps being deplatformed from Amazon or one of its competitors, loss of the trust of readers and reviewers, and--no doubt--more.
Moreover, the better the author, the more likely he or she is to be falsely flagged!
For school children and university students, there is the risk of accusation, punishment, bad grades and shaming if teachers happen to use a bad AI detector that is unreliable. Parents should be aware that false positives happen, and not all AI detection products are equal (or trustworthy).
There's probably no appeal or argument for a college or university applicant whose essay is run through a bad tool.
The Authors Guild ran a test of five of the "top" AI detection tools using ten vintage Authors Guild articles, all of which were written in 2020, before AI technology existed. Three products performed well. One was utterly unreliable.
"AI detection tools are AI models trained to recognize statistical patterns associated with large language model [LLM] output, such as sentence rhythm, vocabulary distribution, and predictability of word choice. But polished, edited prose written by experienced human writers shares many of those same characteristics, because LLMs were trained on polished, edited prose written by experienced human writers. The more refined and controlled a writer's style, the more it may resemble the output these tools are designed to flag."
https://authorsguild.org/news/can-ai-detectors-be-trusted/
All the best,
SPACE SNARK™
Posts
No comments:
Post a Comment