AI Has Killed Translation. Again.

The language industry has been facing its own extinction for decades, yet somehow, we are still here.

It all began 72 years ago, in 1954, when Georgetown University and IBM demonstrated a computer capable of translating Russian sentences into English. This major achievement made headlines, leading some researchers to believe that this was just a first step into a promising future. A future where high-quality machine translations would thrive, and human translators would soon become unnecessary.

Since then, technology has made many major leaps and bringing us step by step to where we are right now. By late 1960s, early systems like SYSTRAN were built to tackle technical manuals translation for government agencies, showing early promise but remaining rigid and limited. In 2006, Google Translate made its automated translation freely accessible to millions of people. By 2017, language tools like DeepL raised the bar even higher by producing translations that sounded remarkably natural.

Today, we live in the era of generative AI, where language models such as ChatGPT or Gemini can translate, rewrite, explain, and even adapt entire documents in mere seconds.

It feels like a groundbreaking moment, but for those who have spent many years in the language industry, it carries a deep sense of déjà vu. Once again, the familiar prediction echoes, “We won’t need translation agencies anymore”.

Every technological breakthrough follows a similar pattern.

First comes the wave of excitement, as clients discover something they previously paid professionals to do, can now be done in an instant with no cost. This triggers another question, “why are we still paying for translation?”

Encouraged by these breakthrough, clients start to experiment on their own. Discovering how much technology can accomplish for so little effort, and quite often, they are impressed by the results. Who wouldn’t be? The technology is efficient and hard to ignore.

But sooner or later, a realization sets in, “The translation looks fine, but can we actually use it?”

That’s when things become interesting.

This is where the distinction between simple fluency and true professional utility becomes clear. Understanding a foreign language in an email is one thing, but publishing a maintenance procedure, a medical device manual, or a legal contract is entirely another.

While the translation may sound fluent, critical questions begin to surface. Who checked if the technical terminology is accurate? Who verified the instructions as the manufacturer intended? Who ensures the consistency across dozens of documents and languages?

And when something goes wrong, who takes the responsibility?

Ultimately, algorithms do not take accountability, but humans do.

It would be foolish to pretend that generative AI is just another version of Google Translate. Its capabilities are vastly broader, and its economic impact is impossible to ignore.

Some basic translation work will never return to the way it was. Today, clients expect faster delivery at lower costs without sacrificing accuracy or reliability. For agencies and Project Managers, the challenge isn’t choosing between defending outdated workflows or blindly trusting AI, but finding a way to combine technological efficiency with specialized human expertise.

So, is this finally the end of the industry?

Every time translations becomes easier to produce, we discover new reasons why human expertise matters. Technology will continually advance and play a growing role in our lives, so rather than fearing it, we should work alongside it.

Instead of endlessly asking whether AI will replace translators, shouldn’t we adapt alongside the technology? Our value is shifting away from mere textual conversion toward essential human oversight.

AI can process words, but it cannot exercise editorial judgment to shape a brand’s unique voice across different global markets, nor can it grasp the cultural nuance to catch subtle idioms or local sensitivities before they turn into costly marketing missteps. In highly strict fields like law or healthcare, where a misplaced word can lead to dangerous consequences, automated systems fall short. Above all, AI can’t deliver the ultimate assets, which is trust and accountability. These are critical qualities that technology alone simply cannot replicate.

Agencies that learn to combine technological speed with professional reliability won’t find AI to be the end of their business. In fact, it forces a necessary evolution on how language services providers operate. By shifting the narrative away from survival toward strategic empowerment, the real value of human intuition becomes clearer than ever. In a world flooded with instant, synthetic content, verified precision and human accountability become our true advantage.

This could be the beginning of a more competitive world, but it’s far from over.

After all, we have been hearing about the death of translation since 1954.

Yet, we’re still here.

Inspired by an article from:
Lorenzo Mataloni. “AI has killed translation. Again.”

https://www.linkedin.com/pulse/ai-has-killed-translation-again-lorenzo-mataloni-fekrf/