Act 1: The Snapshot

We begin with our feet firmly planted in the present and reflect on where the language services industry currently stands (spoiler: there’s no way to avoid this thing called GenAI).

But do we understand where we are and where we are headed? Are we asking the right questions, or should we just start with the “Why”? 

There is a sense of urgency that powers the varied market players. We explore new pathways, test new solutions, learn from the mistakes made along the way, course-correct, and start again.

The industry is evolving, and we are in motion. In the following pages, we examine what this movement looks like. 

Charles Dowdell

Manager, Technical Communications at Komatsu

Charles has 25+ years of experience managing technical communications teams and processes. He has significant global experience in multinational corporations in several different industries. He is an experienced conference presenter and enjoys a good laugh coupled with professionalism in Tech Comms and Information Science. His career started as a global field service engineer. He is an alumnus of Syracuse University iSchool and has worked on all seven continents.

People often overlook how significant language translation truly is. Yet, translation and localization are unquestionably at the heart of global commerce.

Global commerce is a living, breathing, interconnected ecosystem of billions of people, spread over considerable distances. Distance creates scarcity. Scarcity creates value. The ability to create value builds trade networks and powers our interconnected world. Language translations in this system are not just a matter of terminology cross-references and grammar – they are inextricably linked to culture. As such, we need to expand our ability to cross both language and cultural barriers to keep pace with new ideas and spread real value. Efficiency in communicating new ideas benefits everyone.

Enter the era of generative AI tools and processes. While AI has existed for decades, we now have the computer speeds, affordable mass storage and access to unimaginable amounts of content to provide transformative potential. Add to this the societal push to embrace new technology, and we now have tools capable of condensing decades of study and experience into actionable insights.

It’s worth noting that the underlying technology of modern GenAI processes originated in the language services industry. Stiff competition, razor-thin margins, and rapid turnaround times have created the perfect crucible for change and automation.

At Komatsu, we’ve seen this firsthand. We recently took large amounts of structured content that had been translated into English – specifically Japanese English which, while technically English, wasn’t fully optimized for North American users. Using the standards set out by ASD-STE100 Simplified Technical English and other training ingredients, we rewrote that content with the help of modern technology on a large scale. 

This project got us thinking…if we can do automated localization of language with this kind of depth, maybe we can do an even finer-grained localization.

Localization has traditionally been segmented by geographies (EN-US is for American English, EN-GB for British English, and so on). But what if we micro-segmented the localization aspect, training on the domain language rather than the geography?

After all, that’s exactly what technical trainers do. They take the language chain to the last step by interpreting the manual and translating its meaning specifically for the end user, intertwining cultural subtleties familiar to the domain and the user.

Different knowledge domains – whether physics, medicine or software engineering – have their own unwritten rules of grammar, word order, phrases and terminology. For example, the term “bandwidth” has a domain definition in physics as well as popular culture and broadly describes human effectiveness, too.

Imagine if we could expand the role of language localization to rewrite instructions based on what the user digests best, understanding both the user’s domain experience and their learning style.

Using GenAI and automation assists in breaking down the walls of localization, possibly allowing for training a model down to a specific user. A help topic explaining a software feature could be rewritten in the style most approachable to that specific user. A medical instruction could adapt its terminology to a patient’s level of understanding without losing accuracy. With GenAI, we could potentially capture the language of the user through speech-to-text and train the models on an individual’s unique local language.  

Yes, the manual exists in English, but whose style of English is it? Is the content presented in an approachable way that is reminiscent of my upbringing and unique experiences?

There is an unexplored frontier of using AI to 

provide content localized to the domain and individual user level.

It‘s often argued that a translated text cannot be definitively “better” than the source language text it derives from. The input from the original author is indeed finite. However, when we add the vast and available external context that the machine can now process, the target will be enriched to a scale that bridges domains, enhances understanding and unlocks value. If we can translate content from the language of design engineering to the language of service technician, we aren’t just converting words – we’re building understanding and spreading knowledge.

Read the full 132-page Global Ambitions: (R)Evolution in Motion publication featuring vital perspectives from 31 industry leaders on the ongoing AI-spurred (r)evolution.

Get your digital copy today