Do our clients trust AI or us?
Client engagement was one of the three themes at EX:CHANGE 2026 for a good reason; we can no longer expect that clients will automatically turn to professionals when they require translation or interpreting services. But how do we position ourselves in relation to machine translation and AI? Christophe Fricker offers a view about the importance of building trust.
If we want to reboot our client engagement, we need to know why so many non-linguists use machine translation tools, including AI chatbots, and we need a clear and effective response. A client told me recently: “I really do want to work with human translators. I know that they deliver something that is far superior than what I can get out of an AI tool. But the experience of working with a translation agency is the same as putting something into an AI tool. You submit a text, you get one back, it just takes longer.” My interlocutor says she feels neither seen nor heard, and that the excitement of dialogue and engagement is absent in her engagement with the translation agency.
In 2025, KPMG surveyed 50,000 people and found that while 66% use AI regularly, only 46% say they are willing to trust it [i]. This is interesting, but researchers are increasingly sceptical whether surveys can tell us anything about how trust works. Looking at the KPMG survey, we can only conclude that 46% of respondents are willing to say that they are willing to trust AI. Some might, however, assume that 66% are willing to trust AI enough to use it. But this would be a misleading assumption because reliance on technology is not trust.
What, then, is trust?
Trust is a prereflective attitude, developed over a long time, whose paradigm is an interpersonal relationship.
Firstly, prereflective means it is not primarily the calculated assessment of risks and potential gains. This means that increasing the amount of information about you (and your skills) available to a potential client will not automatically make you more trustworthy, or increase their trust in you. The linguistic expertise of our clients is often limited, so the fact that translators enjoy the trust of many of their clients is proof that detailed knowledge about language or the resources we access or the processes we follow is not a necessary component of a successful business relationship.
The second aspect of trust is its relationship to the long term. As a habitualised attitude, trust is experience-based; it has a history. But it is also an enabler; it is future-oriented. It is an attitude that develops and is deepened over time. However, this is largely true for reliance as well, and might partly explain why many people have quickly adopted AI tools. I would like to suggest that the reason why AI-powered chatbots have been adopted so widely and so rapidly is that, from the user perspective, they are not new. In experiential terms, there are many similarities between using an AI chatbot and performing a Google search. Adoption was fast, because reliance was already present.
The third aspect of trust is its interpersonal nature. This should be of interest to translators as our activism has long been focused on visibility as a means to gain recognition and greater reward. Cultivating identifiable styles and voices (our brand), having our names acknowledged (#NameTheTranslator), and achieving a public profile are part of this endeavour, and even as relatively recent phenomena, they address a much longer history of readers wanting to know more about the personality of authors; they show us that ‘the author’ (including the author of a translation) is not dead.
AI proves, rather than disproves, this thesis. Users refer to chatbots as people, sharing secrets with them, being polite to them and engaging with them regularly. It is an intrinsic part of our humanity to feel that the world speaks to us, and to be grateful for it. This does not mean, of course, that chatbots have empathy, or that we should believe they do; but it indicates that we have empathy, and that we respond favourably when this seems to be confirmed.
Why is trust an appropriate focus for a discussion about client engagement?
But chatbot interactions are reliance rather than trust. Trust is fundamentally relational, requiring ‘persons’ on both sides of the equation, with each having some leeway to make decisions. It emerges when everyone feels that they act together, when our normative expectations of the behaviour of others are fulfilled, and when our own behaviour is recognised. Producing a good translation and simply handing it over isn’t enough; if we want clients to value our work, we have to invest in the relationship, not just the output.
This means that we have to go out and start conversations with people who are not like us; with people who may not like us; with people whose understanding of language is minimal; with people who have a lot of expertise in areas that we know little about; with people whose business experience exceeds our own, and to conceive of these conversations as the core of our professional activity. Trust is earned by those who deliver information, manage risk or simply choose the right adjective, but only if they make and keep their promises in their business relationships.
Reference
[i] “Global study reveals trust of AI remains a critical challenge reflecting tension between benefits and risks“ KPMG (2025) https://kpmg.com/xx/en/media/press-releases/2025/04/trust-of-ai-remains-a-critical-challenge.html
This article was published originally in the October 2026 edition of ITI Insight.
