Back-cover blurbs, visuals, emails to literary agents: more than twenty employees tell Wired how AI is becoming established in the publishing industry, without the authors’ knowledge.
Officially, the American publishing industry is standing up against generative AI, with lawsuits for copyright infringement and cancellations of books suspected of having been written with an LLM. Behind the scenes, at least three of the Big Five groups, HarperCollins, Simon & Schuster and Hachette, are integrating these tools into their processes, without publicly announcing it or asking for the authors' approval. This was reported by Wired on October 9, based on anonymous interviews with more than two dozen employees.
"AI champions" appointed ex officio
Claude, ChatGPT or Jasper would be used to write promotional and back cover texts, to create covers, to produce marketing material and even to write to literary agents, some of whom suspected automatically generated rejection letters. At HarperCollins, management allegedly bought Claude licenses before looking for how to use them, and then designated dozens of employees as "AI Champions" to find use cases. Ethical, legal and environmental objections raised during these sessions were swept aside, which employees, often poorly paid, see as an affront.
Hachette says it supports AI for operational tasks but not for creative uses or to communicate with its authors and partners. Simon & Schuster says its teams have a limited number of validated tools, the use of which is not mandatory. HarperCollins did not respond.
Read the full article on the original source
Go to the source →