AI-Enabled Technical Content Delivery – Impulses for Technical Communication

A summary of Kees van Mansom, Technical Publication Services Lead Europe at Accenture. In 1991, he wrote his first technical instruction to prevent cold cracking in welding steel bridges. It was then that he learned to put himself in the user’s shoes and tailor documentation to their specific context and needs. His motivation has always been to help others, designing solutions that positively impact people’s work and lives — like the four-page welding instruction that enabled bridges to stand for over 100 years. More than 30 years later, he builds bridges within organizations, innovating how Technical Publications are developed and used. At heart, he remains a technical writer, combining innovation, storytelling, and clear communication to design and explain impactful solutions. As a leader in technical publications, he helps clients transform their processes through state-of-the-art technology and by challenging the status quo to drive meaningful change in people, processes, and technology.

This article summarizes the IUNTC talk from January 29, 2026, December 4, "AI-Enabled Technical Content Delivery – Impulses for Technical Communication". AI-enabled technical content delivery has the potential to be a true game changer for technical publications. In this talk, he explores how AI can create more contextualized and immersive user experiences — while addressing challenges such as trustworthiness, relevance, and accuracy. He highlights the crucial role technical communicators play in preparing content and shaping AI agents for reliable, targeted content delivery.

Read full article

The recent IUNTC meeting was exceptionally well attended and clearly demonstrated how relevant and timely the topic of AI-enabled technical content delivery currently is for the field of technical communication. The high level of participation and strong interest show that many organizations are actively seeking guidance on how to integrate artificial intelligence into their information processes in a meaningful, sustainable, and economically viable way.

We were pleased to welcome Kees van Mansom as a speaker on this topic. Kees is Technical Publication Services Lead Europe at Accenture and brings extensive experience from international industrial projects, particularly in aerospace and defense, automotive, and manufacturing. His perspective is strongly shaped by practical experience. Early in his career, he observed that even high-quality documentation is often not used if users can obtain information more quickly in other ways, such as by contacting experts directly. This insight leads to a central question that is more relevant than ever: How can technical information be delivered in a way that truly supports users at the moment of need?

AI-enabled technical content delivery can be viewed along three central process dimensions: content delivery and user experience, content architecture, and content creation and updates. Together, these dimensions form the framework for a sustainable transformation of technical communication.

Content Delivery and User Experience

The way technical information is delivered is undergoing fundamental change. Traditional documentation formats such as extensive manuals or static PDFs no longer meet the requirements of many modern work environments. In service, maintenance, and operations, there is an increasing need for situational, context-driven support. Technicians and engineers require precise, relevant information at the exact moment they perform a task.

In practice, a significant portion of working time is still spent searching for the right information. At the same time, organizations face growing pressure to capture the knowledge of experienced employees and make it accessible to new generations. As a result, user experience is becoming a decisive factor. The goal is to integrate technical information more closely into operational workflows and close the gap between documentation and physical action.

Artificial intelligence enables concrete solutions that are already being tested and deployed. One example is the use of assistive systems combined with mobile or wearable devices. During maintenance activities, technicians can receive context-specific instructions through headsets, tablets, or embedded applications. Sensors, cameras, or acoustic analysis can assess the current condition of a system and automatically retrieve the most relevant procedures.

These solutions typically combine several capabilities:

automated identification of components or failure patterns
retrieval of relevant instructions from structured content repositories
step-by-step guidance during task execution
real-time feedback through voice, images, or video
documentation of the performed work.

An additional benefit lies in continuous feedback. Users can report missing or unclear information directly while performing tasks. This feedback loop enables ongoing improvement of content quality and usability.

These developments illustrate that AI should not be understood as a replacement for documentation, but as an interface that transforms existing knowledge into interactive, context-aware support.

Content Architecture

A robust content architecture is a critical success factor for AI-enabled content delivery. While many organizations invest heavily in AI platforms, the structure and quality of their content often receive insufficient attention. In practice, the greatest impact is achieved through systematic modernization of the information foundation.

Unstructured and inconsistent content frequently leads to incomplete or inaccurate AI-generated results. A sustainable solution therefore requires a transition toward modular, standardized, and semantically enriched content.

Concrete solution approaches focus on multiple areas. In long-term transformation programs, structured information models are introduced, such as standardized topic or data module frameworks. Content is enriched with metadata and closely linked to product and engineering data. This enables AI systems to identify and retrieve relevant information with greater precision.

However, transforming the entire authoring landscape is for many organizations a bridge too far. In addition, such a transformation takes time and will not cover all legacy content. In those cases it makes sense to first implement a transitional solution. One such approach is the use of aggregation layers. These platforms consolidate content from different sources, including PDFs, HTML documents, knowledge bases, and databases. AI-driven analysis is used to segment, classify, and semantically index this content.

In practice, this may involve:

automatically breaking down existing documents into topic-based modules
recognizing structural elements such as headings, procedures, and safety information
proposing or generating metadata
enabling semantic search and AI-based retrieval.

This approach allows organizations to deploy context-aware solutions without immediately replacing existing systems. At the same time, it creates a foundation for gradual migration toward structured and future-ready content environments.

Another emerging solution is the stronger integration of technical documentation with engineering and service processes. By linking content to product configurations, variant management, or bill-of-material data, instructions can be dynamically tailored to specific systems.

This architecture not only enables scalable AI applications but also reduces complexity, redundancy, and maintenance effort over time.

Content Creation and Updates

The creation and maintenance of technical content are also evolving through artificial intelligence. While public discussions often focus on full automation, industrial practice shows that the greatest value lies in collaborative and assistive approaches.

A key concept is co-authoring. In this model, AI is used to support authors in different phases of the content lifecycle. This is particularly effective in knowledge capture.

One practical solution involves systems that record and analyze interviews with subject matter experts and generate structured draft content. These drafts already include typical elements such as prerequisites, step-by-step instructions, and expected results. The draft serves as a basis for immediate validation and refinement. This significantly reduces the time required to transform expert knowledge into usable documentation.

Another important area is change management. Engineering changes often lead to extensive manual effort in updating documentation. AI-based solutions support this process by analyzing change requests, identifying affected content, and proposing updates.

Typical capabilities include:

semantic search to identify impacted modules
automated suggestions for updating text, values, and references
generation of new content based on existing patterns
consistency checks across documents
visualization of impacts across product variants.

In practice, these functions are increasingly integrated directly into authoring environments. Authors continue working in familiar systems, while AI provides background support and automates repetitive tasks.

In the longer term, these solutions are evolving toward orchestration. AI will not only support individual tasks but will help coordinate entire workflows. Change requests can be analyzed automatically, impacts simulated, and affected areas identified across systems.

This shift leads to a stronger strategic role for technical communication. Technical communicators increasingly act as information architects, process designers, and quality managers.

Conclusion

AI-enabled technical content delivery represents a comprehensive transformation of technical communication. The greatest value is not achieved through isolated technology initiatives, but through the integration of user-centered design, structured content architecture, and collaborative workflows.

Concrete solutions demonstrate that this transformation is already underway. Assistive systems provide real-time support for users, aggregation platforms unlock the value of existing content, and co-authoring models improve efficiency and quality.

Organizations that successfully integrate these three dimensions create the foundation for more efficient processes, higher quality, and a better user experience. Artificial intelligence will be most effective as an assistive and orchestration technology. However, the foundation remains high-quality content, consistent structure, and a deep understanding of user contexts.