Our Analysts' Insights

Here is the original recipe for the coming labor apocalypse: Take more than 2,000 work activities for more than 899 occupations. Stir in the amount of time spent on performing those jobs. Determine the technical feasibility for automating them. Yield: Potentially eliminate 49% of the American workforce that currently performs those tasks. Adapt the ingredients for your country to create your own existential crisis soufflé.

Much of the mainstream business press coverage of automation and artificial intelligence begins with that basic recipe, originally concocted by McKinsey & Company in 2017, and applies it to various industries, professions, and countries. Language services often show up as one of the victims in such analyses. Pundits predict that AI and its neural machine translation (NMT) spawn will obsolete an entire industry and put hundreds of thousands of people out of work. That could happen – but only if we postulate a future where the language sector stands by and does nothing. Inaction will result in the wholesale annihilation of many providers, but CSA Research has observed enough tech-enabled LSPs where it’s not business as usual to be more optimistic.

What will success involve? On a corporate earnings call earlier this year Microsoft CEO Satya Nadella said that, “Digital technology today is not about tech companies doing innovation. It is about the rest of the world doing innovation with technology.” For much of the world, that revolves around initiatives such as digital transformation, customer experience, and ubiquitous access. Enterprises around the globe are hard at work on projects that they hope will generate more revenue, develop new markets, and build stronger customer relationships, wherever and in whichever language they do business. LSPs need to get closer to the core of that shift in business (“The Business of Language and Evolution of LSPs”).

Some language service providers have taken notice. While LSPs have traditionally supported global expansion, the innovation that Nadella references drills even deeper – and has caused visionary LSPs who are paying attention to rethink their own roles in the context of five factors:

• Digital transformation crosses many borders, including language. The ongoing digitalization of industry and government has led to massive changes and growing demand for operational efficiency (“Global Digital Transformation: Why Firms Fail”). Global communication is essential to this shift. While NMT threatens the status quo in the written-word translation market, LSPs bring skills such as system integration, domain area expertise, understanding of market complexity, and inputs for machine learning. They work with NMT and human translators to make information accessible in multiple languages and the formats that are engorging the world – multimedia, non-textual, over-the-top streaming services, online gaming, spoken-language, and conversational interfaces are essential. CSA Research expects growing activity and revenue as Global 3000 and medium-size enterprises increase investment in global digital transformation (GDX) and customer experience (GCX) strategies. We also predict increasing revenue going to those individual linguists who can adapt to provide the needed services.
   
• Burgeoning adoption in digital channels drives more language requirements. Massive growth in markets such as digital-everywhere China and video-first India underscore the need for global enterprises to offer great experiences in a growing number of markets. Some consumer-facing companies that we’ve advised plan on increasing their portfolio from 30 to 60 to 100 languages in the coming years as they strive to reach additional tens of millions of Indians in local languages, people across Africa, and lesser-served populations around the world. Spoken-language interfaces for marketing, voice assistants, and advanced conversational interfaces raise the stakes even higher. Perfecting speech and interpreting for these devices presents a Zeno-like paradox to companies such as Apple, Google, and Samsung – machine-generated written and spoken translation will never be perfect, and the number of languages needed for outstanding customer experiences will continue growing.
   
• Empowering technology is increasingly available at reasonable prices. As much as their translation revenue prospects may suffer from NMT advances, companies in the language sector benefit from mainstream corporate investment in and academic research on cloud services, micro-services, machine learning (AI), big data, content intelligence, and natural language processing (NLP). These technologies support a range of capabilities such as process automation, textual analysis, semantic nets, neural machine translation, and sophisticated supply chains. As tech-savvy LSPs integrate these facilities, they can optimize processes, offer more services, and scale their operations to meet challenging client demands. Importantly for LSPs, these technologies don’t work right out of the box. They need to be trained, integrated, applied to the right type of content at the right time for the right target audience in the right applications. Until AI and NMT become self-describing, self-installing, self-operating automata, LSPs can position themselves as those essential integrators – but even if Ray Kurzweil's prediction of 2029 as the year that AI passes the Turing test is correct, humans will still be training machines to improve evolving languages.
   
• Content creation, processing, and transformations challenge everyone. No matter the country, industry, or business function, everyone we interview about global content acknowledges that they’re not doing a good job managing it. They need help – and some full-service LSPs are gearing up to provide it. Their core services of translation, interpreting, or localization have given them insight into the information flows of their clients. Some of them are looking upstream for more strategic roles not only adapting content for international markets, but also taking on BPO responsibility as global content service providers (GCSPs – “Who’s Who in Language Services and Technology). 
   
• Outside investors seek bigger multipliers. Over the last few years private equity groups and venture capitalists have invested in the sector, some buying large and medium-sized LSPs that are well-positioned to extend their capabilities to support global digital transformation and customer experience initiatives. These investors will give the enlightened LSPs enough to flesh out their product and service portfolios with the right combination to meet that changing demand.

The existential threat to the language sector is real. So too is the change that some LSPs are making in response. Looking forward, it’s not a binary issue of “Will AI  and NMT destroy the language services industry by 2025?” Instead, the relevant question is, “What will the language services sector look like in 2025?” Those are two very different discussions that we regularly have with both suppliers and buyers.