AI Working Group

Scope of this effort

In book publishing, applications of AI can be found in editorial creation and management, metadata creation and management, rights management, accessibility, and forecasting. The most widely discussed use cases focus on evaluating content, generating content, and copy-editing or proofreading content. However, promising supply-chain examples can be found in AI’s ability to analyze larger data sets, strengthen sales forecasts, and evaluate market trends.

Concerns about the use of AI in book publishing include inadequate controls around use of copyrighted material, inaccurate quality control for data used by AI engines, the quality of data scraped to inform large-language models, and the reliability of AI outputs. In addition, the need to disclose to consumers when AI is used to create content has not been uniformly agreed upon in publishing or the broader marketplace. Because the field is changing rapidly, publishing organizations also face challenges keeping up.

Current State

Although aspects of artificial intelligence (AI) and machine learning date back decades, practical applications jumped into industry and public consciousness in the fall of 2022. Since then, a growing number of companies have announced AI-related products. Throughout 2024 and early 2025, many larger publishers reported the use of AI in one or more parts of their businesses. 

Early applications addressed marketing and editorial workflows, but the pace at which licensing for AI is moving forward is also accelerating. Other supply chain examples are likely already in development. More news is expected in topic areas like metadata and forecasting, currently part of BISG’s supply chain committee efforts. Within book publishing, we have yet to see the emergence of a pan-industry review of the impact of AI on our business, but that does not mean such uses are not being considered.

Objectives

• Collect publicly available information on book-publishing-related applications of AI, evaluating applications to determine their potential impact. Consider the need for labeling for either content generated by AI or use of AI in product development.

• Lead or facilitate discussions about effective and concerning AI applications

• Propose a policy or a collection of best practices that support the use of AI while addressing issues like intellectual property copyrights, privacy, and bias

• Publish findings as appropriate

Stakeholder Impact/Benefits

• Gathering information about current applications can increase awareness and understanding of AI

• Examining AI applications in editorial, metadata, rights, accessibility, and forecasting provides the basis for potential breakthrough developments that would improve efficiency or effectiveness across the supply chain

• Understanding risks accompanying the use of AI in book publishing environments 

• Engaging multiple committee efforts provides an opportunity to reduce the siloed nature of current discussions about metadata, rights, and supply chain issues

Deliverables

• Identify and share non-BISG resources that provide examples of AI applications

• Propose policy or  industry best practices related to the use of AI

• Offer an eight-part webinar series that helps the industry “see around corners” with respect to the use of AI across the supply chain.

Blockers

As is often the case with committees and working groups: Representation in key areas needs to be maintained so that the committee can fully address issues. The current workflow committee lacks significant depth in this topic area. The working group will continue to solicit engagement from BISG’s full membership to improve its ability to recruit and engage subject matter experts from across the supply chain.