The perceived downward trend in translation quality and linguist working conditions since the rise of machine translation (MT) and generative AI (GenAI) — as reflected in numerous translator surveys — is causing a growing number of professional associations to take an official stand against relying too much on these technologies. Most recently, France’s Société française des traducteurs (SFT) and the UK’s Institute of Translation and Interpreting (ITI) have published statements outlining the multifaceted risks of over-reliance on AI.

These manifestos echo the stances already taken by a variety of language organizations, from those focused specifically on creative translation (such as Audiovisual Translators of Europe) to those advocating for translation professionals in general (such as the American Translators Association (ATA)). The associations’ statements raise several shared concerns, which speaks to their unified outlook on the future of the industry’s employment of AI.

Translation quality risk factors

As maintained by the ITI, human translators “bring nuanced understanding, creative flair and an innate feel for their working languages, honed through years of experience and specialisation.” Adversely, as stated in the SFT’s manifesto, “algorithms do not have the capacity to understand, question and take account of context.” The consensus is clear — while AI can be a useful tool and increase productivity, it cannot replace the multidimensional perspectives and cultural sensitivity of human translators.

Both manifestos, as well as those published previously, emphasize the importance of accuracy and quality in translation, which often requires a slower, more deliberate process than AI can provide. The ITI thus calls for a “slow translation movement,” stating that “rushed translation can lead to errors[,] unnatural language and the loss of crucial subtext.” The SFT outlines similar risk factors: “awkwardness, semantic changes, even contradictions or omissions,” as well as the production of “partially or wholly false information.”

Impact on clients and language professionals

The organizations also highlight ethical concerns for professional standards in language services. The ITI maintains that “professionalism and professional ethics must always be prioritised above shortcuts,” such as using AI with no human input, while the SFT warns of the outcomes of such practices: “forgoing confidentiality and facing considerable risks: legal, economic, liability, security, reputational, manipulation; as well as error risks.” If unchecked, AI-led translations can have an unintended impact on the quality clients aim to produce, as well as the security of their businesses.

Finally, both manifestos and their predecessors caution all industry stakeholders on the professional jeopardy translators and interpreters face as GenAI becomes more prevalent. In response to the current state of affairs, “the SFT denounces the lowering of remuneration and the degradation of working conditions for highly-qualified professionals, even while the global market continues to grow.” Similarly, the ITI emphasizes that the level of human translator competence “deserves to be valued and appropriately remunerated,” hinting at the issue of decreasing standards, both quality- and compensation-wise.

AI as a tool, human translators as the priority

Apart from analyzing and highlighting the value of human-first translation and the risks AI brings to clients and translators alike, the associations suggest a new way of aligning tech with the work of human translators. With its emphasis on “slow translation,” the ITI advocates for “quality, responsibility and accuracy above speed,” thus underscoring the need for human translators in the production of quality multilingual content.

The SFT again reflects this stance, recommending that GenAI be “viewed as an assistant” rather than a standalone solution. In addition, the organization calls for the recognition of human translators “as critically minded experts” capable of assessing the usefulness and price value of AI-assisted translation, transparency in disclosing the origin of AI-generated content, and the reversal of the downward trend in pricing, with an eye towards “improving the entire ecosystem” of the industry. Critically, SFT also directly calls for governments to prioritize human translators and to refuse funding GenAI-led production.

Perception of AI in highly specialized translation

While SFT’s survey respondents show concerns in line with their manifesto, not all translators are feeling the heat of AI quite as much. Chris Durban, a French-to-English translator specializing in finance and business, views SFT’s statement as “an attempt to speak to the concerns of members in less specialized areas, some of whom are feeling real pain. Many of them work through agencies that are also under severe pressure. So it’s a cri de coeur in a part of the market where the power dynamic is very skewed.”

In her specialized market, Durban doesn’t perceive the same pressures: “I’m just back from ASTTI’s summer school for financial translators, which featured high-level presentations by professionals in finance, regulatory environments, and the like. There was little pearl-clutching over AI that I could see, because the more specialized the market, the higher the stakes. And the more obvious it is that human expertise is essential. We craft high-profile documents for high-risk situations; AI is another tool in the toolbox. Professionals have to know how and when to use it. Given the hype, it’s natural that clients test AI ‘solutions’ in certain cases — the tech bros are out there, selling relentlessly. But the challenges in my markets are different, and clients are ultimately pretty risk-averse.”

With associations sounding the call for a more balanced approach, as well as the growing number of studies on the discrepancies between human and AI translation, it does not seem the industry is at a point of no return. Rather, it is learning to adapt to the new technological landscape and manage its risks. The combined efforts of advocating bodies, researchers, and industry stakeholders who continue to view human translators as indispensable present a sustainable course towards a future in equilibrium.