By Fardous Bahbouh, Researcher & Consultant on Labour Rights, Public Policy, and the Political Economy of the Translation Industry

The translation and interpreting industry, and some academics, keep talking about adaptability, flexibility, and collaboration. But these buzzwords often mask a deeper problem: exploitation. What we need is fairness, not compliance to algorithms or precarious systems.

In my research, I see how AI, algorithmic control, and surveillance are already reshaping the industry — and not for the better. Translators and interpreters are under pressure to stay “on queue,” constantly checking devices and emails, modifying their working habits, and competing for jobs allocated through opaque algorithms. Rates are meagre: public service interpreters in the UK are paid as little as 18–19 pence per minute — and that only counts the actual seconds spent interpreting, not the unpaid waiting time.

This is more than an economic issue. Translators and interpreters are a highly racialised and feminised labour force, already facing structural obstacles to more secure work. Without government action, these dynamics risk increasing in-work poverty, inequality, and even child poverty.

AI isn’t taking translators and interpreters’ jobs — yet. But it is enabling companies to extract more labour for less pay, erode autonomy, and obscure accountability. Platforms monitor keystrokes, search times, even bathroom breaks, feeding that data back into job allocation and pay decisions. It’s a “fast fingers booking system” that keeps skilled professionals in a permanent state of alert.

The result? A workforce tethered to devices 24/7, without security, autonomy, or fair compensation. Even as translation and interpreting companies profit, workers’ livelihoods are increasingly precarious. This is the modern-day sweatshop in an advanced knowledge economy, not in a hidden factory overseas. Freelancers are treated as “dependent contractors” with none of the protections employees deserve — and UK courts are now seeing cases similar to those against Uber and Amazon, challenging these conditions.

The Shifting of Risk

By “shifting more risks onto translators and interpreters,” I am referring primarily to the transfer of risks that would normally be borne by an employer onto individuals who are classified as self-employed and increasingly managed through AI-enabled platforms.

First, translators and interpreters absorb all employment-related risks. As self-employed workers, they are not entitled to minimum pay rates, guaranteed hours, sickness pay, or paid waiting time. Interpreters and translators often have to log into apps and remain available for long periods without pay, waiting for jobs that may or may not materialise. Interpreters in outsourced public service interpreting experience unacceptable health and safety risks: assignments are allocated through platforms with limited information, including no advance disclosure of health and safety risks, such as working with people who are suffering from contagious diseases. During the COVID-19 pandemic, interpreters were classified as frontline workers and expected to continue working, often without adequate sickness leave or protections.

Second, interpreters bear significant financial risks linked to cancellations and unpredictability. For example, court interpreters may have already paid for transport and childcare, only for hearings to be cancelled at short notice. Compensation is often capped at around £50, which may be less than the interpreter’s out-of-pocket costs, meaning they operate at a loss.

Third, some companies impose punitive financial penalties. Interpreters may be fined for lateness or inability to attend a booking, with deductions of around £50 even when the total value of the assignment is lower than the penalty itself. This shifts operational risk and contingency costs directly onto workers.

Fourth, there is the risk of non-payment if companies cease trading. When agencies collapse, translators and interpreters can be left unpaid for work already completed, as seen in cases such as Debonair Languages and Pearl Linguistics.

AI-enabled systems allow companies to scale rapidly while insulating themselves from liability, leaving individual workers exposed when firms fail. Companies reduce their own exposure to employment, health, and financial risks, while systematically transferring those risks onto translators and interpreters who lack job security, bargaining power, or meaningful protections.

Ethical Responsibility of Universities and Professional Institutes

Universities and professional institutes have an ethical duty, too. Collaborating with industry actors without due diligence risks amplifying market fundamentalist views that entrench exploitation and reward companies lobbying for government contracts, even when these companies treat linguists as disposable. Academic and professional credibility should not be used as a rubber stamp for exploitative practices. Partnerships must be scrutinized to ensure they protect workers’ rights, promote fair labour practices, and do not give reputational capital to companies undermining the very values we claim to uphold.

The Solution

Stronger labour protections and regulation are urgently needed. AI and algorithms are not the problem themselves; the absence of rules is. We need:

1. Transparency in how assignments and pay are determined
2. Stronger labour rights and a reconsideration of the freelance model
3. Limits on intrusive surveillance

Without urgent government intervention and responsible oversight by academic and professional institutions, AI and algorithmic management will continue to intensify exploitation, stripping workers of dignity, security, and the value of years of training and education. Fairness — not buzzwords about adaptability or complicit collaboration — must become the foundation of the translation industry.

Note: For a fully a fully referenced and theoretical analysis, please check my previous article reflections build on arguments developed more fully in my article On the Ethical Risks of Translation Academics “Collaborating” with the Industry Without Adequate Critical Engagement and Objective Analysis, which provides a fully referenced and theoretical analysis.

Disclaimer: This critique reflects the author’s independent analysis and does not represent any institutional position. It is not connected to, commissioned by, or endorsed by the University at which the author is completing doctoral studies.