Key points
- Employees at three major publishing houses report unauthorized AI usage.
- Large language models generate back-cover copy and promotional materials.
- Publishers use AI for cover art creation without explicit author consent.
Employees at three major publishing houses report that large language models are being used to generate marketing materials without author consent. According to Adam Morgan at Wired, staff members at HarperCollins, Simon & Schuster, and Hachette have observed this practice. The report indicates that publishers are quietly integrating these tools into their standard workflows for publicity and design tasks, bypassing traditional approval processes that would typically involve the rights holders.
By the numbers
- Three major publishing houses: HarperCollins, Simon & Schuster, and Hachette are identified in the report.
- Dozens of staff members provided accounts of the unauthorized AI usage practices within their respective organizations.
- One primary source: Adam Morgan of Wired compiled the report detailing these internal operational changes.
What the figures mean
The involvement of dozens of staff members across three distinct industry leaders suggests that this is not an isolated incident but a systemic operational shift. According to the Wired report, the use of AI extends beyond simple text generation. Workers indicate that large language models are actively creating visual assets, including cover art, which represents a significant change in how intellectual property is handled. This practice occurs without the explicit consent of the authors, raising questions about copyright ownership and the ethical boundaries of automated content creation in the publishing sector.
What to watch
Security and legal teams should monitor how publishers define "work product" when AI is involved in the creative process. The lack of author consent for cover art and back-cover copy creates potential liability for intellectual property infringement. Organizations should track whether these practices are codified in new employee handbooks or remain informal directives. The scope of AI usage appears to cover both textual publicity materials and visual design elements, indicating a broad integration of generative tools into pre-production workflows that bypass traditional rights management checks.
What to do and how to stay safe: HarperCollins
- Review publishing contracts for clauses regarding the use of generative AI in marketing and design.
- Monitor internal communications for unauthorized deployment of large language models in creative departments.
- Audit existing marketing materials to identify content generated by AI without proper rights clearance.
- Establish clear protocols requiring author consent before using AI for cover art or promotional copy.
Step-by-step guide: Detect Shadow AI: Network Logs, Process Trees and Hidden Signals
General security guidance from the Sandbox Gazette newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
Which publishers are using AI without author consent?
Staff at HarperCollins, Simon & Schuster, and Hachette report that their employers are using AI for marketing and design without author approval.
What specific materials are being generated by AI?
According to the report, large language models are used to create back-cover copy, cover art, and other publicity materials.
Is this practice authorized by the authors?
No, the staff report indicates that these AI-generated materials are produced without the consent of the authors.
