Introduction
Digital Publishing has arrived as a business case in mainstream publishing houses. Still, a large share of the digitally available documents are not or only on a very shallow level accessible. The reasons for the lack of accessibility are variable: lack of knowledge on how to produce accessible documents, lack of tools for making certain types of content fully accessible, or workflows within companies that hinder the efficient production of information in different output channels (print, eBook, web, audio). In addition, the accessible eBook production process is highly versatile and requires many different steps and procedures.
In recent years, Artificial Intelligence (AI) and Large Language Models (LLMs, such as GPTs) have emerged as transformative technologies in the publishing industry. They offer new opportunities to support accessibility at scale. AI-driven tools can automatically generate alternative text for images, simplify complex language into easy-to-read versions, and provide real-time translation into multiple languages. LLMs are increasingly capable of restructuring non-linear content for linear access, assisting readers with cognitive disabilities, and creating multimodal outputs such as synchronized audio descriptions or tactile graphics metadata.
Moreover, AI can optimize production workflows by automating repetitive tasks, detecting accessibility gaps in source documents, and suggesting corrections in line with international standards. This not only accelerates the production of accessible materials but also reduces costs and broadens the reach of inclusive publishing. Importantly, these technologies can be integrated into existing editorial and design environments, making accessibility a natural part of the publishing process rather than an afterthought.
The integration of AI into accessible publishing also raises new questions: How can we ensure transparency and trust in AI-generated content? How do we balance automation with human oversight to respect the rights of authors and publishers? And how can we design inclusive workflows that leverage AI without excluding marginalized voices? These questions highlight the need for interdisciplinary research and collaboration between technologists, publishers, accessibility experts, and end-users.
All the above-mentioned needs demand more and intense R&D to make the process of eBook production more efficient, faster, and the products better accessible and usable. With AI and LLMs now part of the publishing landscape, the potential to create truly inclusive digital content has never been greater.
Call for Contribution
Different formats, types and structures of information (graphics, images, etc.) and the different levels of complexity pose a great challenge in the accessible publishing process. Delivering eBooks and digital content to end-users must be targeted in this Special Thematic Session (STS). Since the variance in the target group is very high (print-disabled, cognitively disabled, hearing-impaired persons), it is challenging to deliver the right content in the right form to the right reader. To implement and reference all the different representations of information should be a field of scope in this STS.
In addition to established topics such as software/plugins, standards and their application, production procedures, automation of processes, workflows, formats, distribution, OCR optimization, accessible graphics, cataloguing/referencing systems, service provision and training, we now explicitly invite contributions that explore the role of Artificial Intelligence (AI), Large Language Models (LLMs), and GPT-based systems in accessible publishing.
Relevant questions include:
- How can AI support the automatic generation of accessible formats (alt-text, easy-to-read versions, multilingual translations)?
- What are best practices for integrating LLMs into editorial workflows while ensuring transparency, trust, and respect for authors’ rights?
- How can GPTs help restructure complex or non-linear content for linear access, benefiting readers with cognitive disabilities?
- What are the ethical and legal implications of AI-driven accessibility tools in publishing?
- How can AI enhance multimodal access (audio, haptic, tactile graphics) and metadata management for inclusive distribution?
We are considerably interested in research results, prototypes, and case studies that deal with these challenges and opportunities. Contributions may address:
- AI-based tools and plugins for accessible publishing
- Standards and guidelines for trustworthy AI within the publishing workflow
- Automation of accessibility checks and corrections
- Integration of LLMs into production workflows
- User-centered evaluation of AI-driven accessibility solutions
- Future perspectives on inclusive publishing in the age of AI
This Special Thematic Session is organized to tackle the above issues in an inclusive way, combining traditional accessibility approaches with emerging AI-driven innovations. Contributions to the STS have to be submitted using the standard submission procedures of ICCHP26. When submitting your contribution, please make sure to select the right STS from the drop-down list “Special Thematic Session.” Contributions are evaluated by the Programme Committee of ICCHP-AAATE and by the chair(s) of the STS. Please get in contact with the STS chair(s) for discussing your contribution and potential involvement in the session.
This STS is organized in cooperation with the Austrian Association BookAccess.
Chairs
Reinhard Ruemer, BookAccess