In this week’s edition: AI developers can’t tell AI-generated text from text written by humans, an Arkansas law concerning “obscene” books takes effect, how the National Braille Press makes books available to blind readers, alternatives to Google Docs, and more.

OpenAI can’t tell if something was written by AI after all
The Verge
OpenAI has decided to discontinue a tool designed to differentiate between human and AI writing due to its low accuracy rate.

Arkansas libraries revise policies in anticipation of law on ‘obscene’ content taking effect
Arkansas Advocate
Librarians across Arkansas are preparing in different ways for a new state law set to take effect next week that changes how libraries handle controversial material. One library system is preemptively planning monthly visits to a courthouse to handle book challenges.

How the National Braille Press Brings Books to Blind Readers
Publishers Weekly
Blind readers engage with braille text in both print and electronic form, in addition to audiobooks, text-to-speech, large text, and other modes of reading. The National Braille Press publishes popular books in braille as well as its own books by blind authors.

Married to the mob: the rise of the smartphone in fiction
The Guardian
Authors are starting to embrace smartphones as key plot points.

Is Google Docs Safe for Writers? Exploring Alternatives
Digital Technology Guru
While Google claims they won’t use your information without permission, obtaining this permission can be easily overlooked.

From Pet Cemetery Owner to Identity Thief to Best-Selling Ghostwriter
New York Times
One writer’s unlikely path to creative success.