The Translation Industry in 2026: What’s Changed in the Last Six Months
It’s been an eventful six months for the translation industry. Between corporate consolidation, layoffs at a machine-translation giant, and a market that keeps growing despite (or perhaps because of) artificial intelligence, one thing is becoming clear: technology is reshaping the tools, but it isn’t removing the need for human translators — especially native, specialised ones. Here’s what’s been happening.
A translation industry that keeps growing
Despite the constant headlines about AI “replacing translators,” the numbers tell a different story. According to Mordor Intelligence, the global translation services market is valued at roughly $64.99 billion in 2026 and is projected to reach $97.65 billion by 2031, growing at an 8.44% annual rate. The drivers are concrete: e-commerce (up to 76% of shoppers abandon a purchase if the content isn’t localised into their language), the boom in streaming and gaming content, and regulatory compliance in healthcare and other regulated sectors, where certified translation by qualified professionals remains mandatory.
DeepL: layoffs and a bet on voice
One of the most talked-about moves this spring came from DeepL, Europe’s leading machine-translation company. In May, it announced it would cut around 25% of its workforce — roughly 250 people — in what its CEO described as a fundamental rethink of the company’s AI-first strategy. Weeks earlier, in April, DeepL had launched a new real-time voice translation suite aimed at meetings, live conversations and frontline customer service, putting it in direct competition with players like Sanas, Camb.AI and Palabra. The irony is hard to miss: even the companies leading machine translation are going through their own turbulence as they try to find their footing in a market that’s shifting faster than ever.
Mergers and consolidation across the value chain
The past few months have brought several notable corporate moves: the merger of Tasq AI and BLEND, combining more than 25,000 experts to bring AI data services and traditional localisation under one roof; Italian firm Aglatech14’s acquisition of fellow localisation company Centrolingue; LeadDesk’s purchase of AI translation specialist Fluentic; and Cineverse’s entry into audiovisual localisation through its acquisition of Giant Worldwide. The pattern is clear: agencies are racing to integrate technology and AI without giving up the human, specialised element that still defines quality.
The human element keeps proving its worth
As machine translation keeps improving, so does awareness of its limits. In literary, legal, medical and marketing translation, cultural nuance, tone, legal terminology and context still require a native professional who understands not just the language, but the culture behind it. Several analyses published this year converge on the same conclusion: the likely future isn’t human translators being replaced, but a hybrid model — AI speeding up repetitive tasks, and native, specialised humans handling what really matters: accuracy, nuance and trust.
September 30: International Translation Day
This year, International Translation Day (September 30) is being marked under the theme “Linguistic Diversity and Language Rights: The Power of Being Understood,” set by the International Federation of Translators (FIT). It’s a fitting summary of where the industry stands right now: in an increasingly automated world, being able to communicate — and be understood — in your own language remains, first and foremost, a matter of rights and inclusion, not just technology.
Recommended video
To explore this topic further, watch “How Big Is the Translation Industry? $71B or $32B?” by wxrks.
Watch on YouTube: https://www.youtube.com/watch?v=ablrGtNKih0
The takeaway
If this half-year has made anything clear, it’s that quality still comes down to people. Technology helps, but for legal documents, medical content, marketing copy, or any project where nuance matters, working with a specialised native translator — with no middlemen or layers of automation in between — remains the most reliable way to get it right.
