The case for optimism 

The most compelling argument from those who scored highly was not simply that AI is bad at languages: it was that quality and discernment become more valuable, not less, in a world flooded with machine-generated content.

As one experienced translator and editor put it:
 

"The more content is machine-generated, the more essential it becomes to have experts capable of evaluating, correcting, refining, and strategically shaping language. In short, automation raises the premium on discernment." 
 

Several respondents highlighted specific areas where human expertise appears resilient. Sworn and certified translation came up more than once, with one practitioner noting that machines cannot accept legal liability; only individuals can. Interpreting also featured strongly, with face-to-face and real-time work seen as particularly insulated from current AI capabilities: 
 

"I believe advanced language skills will not only remain important, but become more valuable in an AI-saturated environment. High-level language work is not about producing grammatically correct sentences; it is about nuance, intent, tone, cultural calibration, rhetorical strategy, and risk management."
 

For linguists working in niche language pairs, the picture was similarly positive. One English-Malay translator observed that AI handles straightforward material reasonably well, but that much of their day-to-day work involves fixing machine translations that are "technically correct but just feel off." Cultural weight, humour, and register remain stubbornly human domains. 
 

There were also signs (echoing the findings of our previous polls) of clients beginning to reconsider their reliance on AI. One freelance translator predicted that when the AI bubble eventually deflates, "there won't be enough skilled professional translators to fill the gap." 

A harder road for some 

The poll's lower scores tell a different story. Several respondents described significant and sustained losses of income. One translator reported losing 90% of their business to AI and unqualified post-editors. Another, with over twenty years of experience, described their turnover falling to a third of its 2023 level, settling just above the minimum wage: 
 

"It doesn't matter what I think; if potential employers are seeking cheap and fast and are willing to compromise on quality, then I can't fight against that. So, it's with a heavy heart that I will move on." 
 

The concern among pessimists is not only about their own earnings, but about something broader: the risk that widespread AI reliance will erode public appreciation of what high-level language skill actually is. 
 

"I think AI/tech has already reduced work in languages because of capitalism, not because language skills are not important. They're just not valued." 

The middle ground 

Among those who placed themselves near the midpoint, the dominant theme was uncertainty rather than despair, and in several cases a clear-eyed view of AI as a tool rather than a replacement. 
 

One interpreter and academic learning assessor noted simply: "Working with AI and knowing how to use AI can enhance our capability." A translator and editor who scored 50 acknowledged that AI is "getting better and better," but argued that humans with good language skills remain essential for checking, editing, and navigating high-stakes fields. 
 

A language teacher offered a different angle: the concern is not just about professional income, but about what happens to language learning itself when people believe they no longer need to develop their own skills. The societal stakes, in this view, go beyond the economics of the profession. 

Looking forward 

What the numbers (and the comments behind them) reveal is a profession that has not lost faith in the value of what it does, even when the market sometimes fails to reflect that value. 
 

The median score of 67.5 is a statement of professional confidence. The gap between mean and median is a reminder that the pain is real for some, even if the majority retain their conviction that skilled human linguists are not going anywhere. 
 

As one member neatly summarised it: "High-level language skills will remain important, although only at a genuinely high level." 
 

CIOL will continue to advocate for professional standards, to support members navigating a changing landscape, and to make the case to employers, institutions, and the public that discernment, cultural intelligence, and linguistic precision are not commodities that technology has learned to replicate. 

Dom Hebblethwaite is Head of Membership for the Chartered Institute of Linguists. For more on Dom see his profile here. 

Views expressed on CIOL Voices are those of the writer and may not represent those of the wider membership or CIOL.