Last Thursday at the Frankfurter Buchmesse, Maarten van Vulpen presented on the Innovation Stage about how structured content and AI together can transform the way publishers create, manage, and distribute content. This well-received talk built on last year’s presentation, Feeding AI Models with Structured Content: The Power of XML ,but this time went further, showing how the connection between structured content and AI is no longer theoretical. Through case studies from scholarly and educational publishing, Maarten demonstrated that when content is structured, AI can finally deliver on its promise: helping publishers get their publications out better and faster.

Why publishing needs a change

Publishing teams today are under constant pressure. They must produce more content, in more formats, for more audiences, all while maintaining impeccable quality and speed. The traditional, document-based workflow simply can’t keep up. Authors and editors move text through disconnected systems, Word files, email chains, design tools, and CMS platforms, introducing delays, errors, and endless version conflicts along the way.

Structured content changes this. By organizing content in a consistent, machine-readable way (for example, through XML), it separates content from presentation and introduces predictability. That means content can be reused, repurposed, and distributed seamlessly across print, web, APIs, and emerging AI-powered channels.

Why AI needs structured content

AI models, however advanced, are only as good as the data they’re trained and fed on. When input data is inconsistent or ambiguous, AI systems struggle, producing inaccurate, incomplete, or even hallucinated results.

Structured content mitigates this risk. Because it is organized, labeled, and enriched with metadata, it provides AI with contextual understanding, a clear picture of what each piece of information represents and how it relates to others. Instead of guessing, the AI can reason. It knows which elements are titles, abstracts, figures, or citations, and how they should be processed or displayed. This clarity dramatically improves AI accuracy and reliability, turning automation into a trustworthy partner rather than a potential source of misinformation.

Scholarly publishing: from manuscript to machine-readable knowledge

In scholarly publishing, XML-first workflows are redefining speed and precision. When a manuscript enters the system as structured content, every part of it, title, authors, affiliations, abstract, references, figures, and tables, becomes machine-readable from the start.

AI tools can then automatically check references for accuracy, validate metadata, flag missing data, and even suggest semantic tags or keywords. Production teams can instantly generate multiple output formats, from journal layouts to online article views and metadata feeds, without manual typesetting or reformatting.

The result: faster publication cycles, fewer errors, and higher consistency across journals and platforms. Structured content ensures that scholarly communication is both human-readable and machine-actionable, enabling not only better publishing, but also better discoverability, reuse, and analysis.

Educational publishing: from learning materials to personalized experiences

Educational publishers face similar challenges, but with a unique twist: the same learning content must appear in textbooks, online lessons, apps, and interactive exercises. All tailored to different levels and markets.

Structured content makes this possible. By managing a single, well-structured source, publishers can automatically generate outputs for different media. When paired with AI, this foundation becomes even more powerful. AI can analyze content structures to generate quiz questions, summaries, and adaptive learning recommendations, helping educators personalize learning experiences without reinventing content each time.

One Fonto customer, for example, now produces both print and digital course materials from the same XML source, with AI-assisted enrichment ensuring consistency and quality across every channel. What once took weeks of manual editing now happens in hours.

Beyond automation: smarter, context-aware AI

When structured content and AI work together, the result isn’t just automation, it’s augmentation. Structured content gives AI systems the context they need to understand meaning and relationships, not just words. It transforms AI from reactive to predictive: capable of identifying trends, suggesting improvements, and even supporting decision-making based on structured data insights.

This approach is also the foundation for responsible AI. By grounding models in high-quality, context-rich data, publishers can reduce hallucinations, enhance transparency, and build trust in AI-driven processes.

Future-ready publishing

As Maarten concluded, “Structured content isn’t just a publishing tool. It enhances AI systems, improves content quality, and provides richer context for smarter automation.”

The publishing industry stands at a turning point. Structured content provides the clarity and consistency AI needs to thrive, while AI, in return, accelerates and enriches structured publishing workflows. Together, they form a powerful combination that drives speed, quality, and scalability.

The future of publishing is already here: structured and intelligent.