AI Translation vs Human Language Skills: Why Organisations Still Need Multilingual Talent

What the decline in language learning means for tomorrow’s workforce and why AI isn’t a get-out-of-learning-free card

Every year, on the 26 September, we celebrate the European Day of Languages. That’s a timely reminder because the data now paints a worrying picture for employers who need people able to work across borders and cultures.

Recent analysis by the Higher Education Policy Institute (HEPI) shows a sharp fall in formal language study: fewer than 3% of A-Level entries were in modern or classical languages in the most recent year, undergraduate enrolments in language and area studies are down about 20% in five years, teacher recruitment is well below target, and strikingly there are now more A-Level entries for PE than for French, German and classical languages combined.

Why This Matters for Employers

Language ability matters for more than literal translation. It sits at the intersection of communication, trust and cultural understanding – three things that affect hiring, negotiation and customer relationships. Employers across sectors consistently report that intercultural competence is a practical, work-critical skill. Researchers and employer surveys show intercultural skills improve team performance, reduce misunderstanding in international projects and are increasingly part of the checklist for graduate recruits and managerial roles.

The Pipeline Problem for Global Talent

When fewer young people study languages, the pipeline of home-grown multilingual professionals shrinks. That raises real operational risks for firms that rely on local language skills for market research, regulatory navigation, partner relations or customer service. It also changes the composition of graduate cohorts: fewer linguists and people with deep cultural literacy enter the talent pool, which makes it harder for organisations to scale international projects without depending on external hires or contractors. The HEPI findings suggest this is not a hypothetical future but it’s happening now.

The AI Translation Debate

It’s tempting to imagine a future where wearables, earbuds or AR glasses provide seamless, real-time translation and render language study optional. Advances in machine translation and on-device interpretation are impressive. But careful reviews and empirical studies show important limitations:

Context and nuance: State-of-the-art systems still struggle with context, polysemy (words with multiple meanings), idioms and pragmatic nuance which are the very elements that determine whether a message is persuasive, polite or commercially appropriate. Machine output can be grammatically correct but socially tone-deaf.

Industry-specific language: Sectors such as law, healthcare, finance, hospitality and technical engineering use specialised vocabulary and conventions. Effective communication in those contexts often requires human judgment, annotation and an understanding of regulatory or ethical constraints that automated systems alone do not reliably provide.

Relationship-building and trust: Speaking a partner’s language, even at a modest level, signals respect and builds trust. That trust converts into faster deals, smoother negotiations and stronger long-term relationships. Technology may transmit words; it doesn’t automatically transmit the cultural competence that underpins durable business relationships. Evidence from employer surveys highlights that organisations value intercultural competence for precisely these reasons.

Operational vulnerability: Reliance on third-party translation tools introduces vulnerability (connectivity, data security, vendor lock-in, or outright failure in sensitive or high-risk interactions). Employers need contingency and in-house capability, not single-point technological dependency.

Practical Implications for Organisations

Treat language and intercultural skills as strategic capabilities, not optional extras. Build them into talent frameworks, graduate pipelines and leadership development. Employers already rank cultural competence as an important employability trait; as the home talent pool shrinks, this will only increase.
Blend technology with people. Use AI and real-time tools to scale and support multilingual services, but pair them with human expertise for quality assurance, culturally sensitive communications, and client-facing negotiations.
Invest early and locally. Apprenticeships, in-company language learning, language pathways for STEM students and partnerships with schools and HE institutions can rebuild the pipeline more sustainably than ad-hoc hiring.
Map language assets to business needs. Not all languages are equally critical for every organisation. Identify the languages and cultural competencies that map to your markets and prioritise those strategically (for example, regional commercial hubs, regulatory jurisdictions, or supplier/partner networks).

Final Reality Check

Technology will continue to improve, and machine translation will be an essential tool in the toolkit. But the idea that translation tech will eliminate the need for people who can communicate across languages and cultures misunderstands the nature of workplace communication. Language learning produces cognitive flexibility, cultural intelligence and relationship skills that machines do not replicate. Employer surveys consistently place intercultural competence among the capabilities they prize most in recruits and managers. In a context where formal language learning is diminishing, organisations that double down on language and intercultural development will find themselves better prepared for complex international work.

👉 At Babel, we support organisations in developing the language and intercultural skills they need to thrive internationally. Explore our Language & Culture Training Programmes.