End-to-end security will become a common requirement for translation. The language service industry’s distributed work model – heavily dependent on freelance translators and interpreters, project managers and testers (on all sides of the table), and everyone else working from home, in cafes, on the beach, or from wherever – is a ticking timebomb. It’s only a matter of time before the industry ends up in the limelight due to a cybersecurity breach.

Mobile device management, multi-factor authentication, and data loss prevention are only a few of the issues that buyers and purveyors of language services and technology must stay on top of. As part of their enterprise-ification mandates, IT managers will push their content management systems (CMSes), translation or interpreting management systems (TMSes or IMSes), and machine translation (MT) suppliers to support an array of security and privacy laws in every market where they do business. A growing number of them will adopt content and translation management solutions that force linguists – and the LSPs they work for – to work within their locked-down, trusted, and secure environments. ISO 27001 certification will become a critical part of the RFP selection process.

LSPs will follow software vendors to subscription selling. Prompted by the move to subscriptions in many facets of their lives, a growing cadre of language service providers including Jonckers, PureFluent, and RWS offer translation – and some interpreting – as a subscription rather than a transactional service.

Their mission is to combine human expertise with automation, simplify the financial aspects of the buyer-seller relationship, and make client revenue more predictable and less episodic. These providers will offer a selection of packages for a varying scale of purchases, together with add-ons such as connectors or specialized services. We expect medium-sized enterprises to lead the charge in seeking a simple solution to their language needs that combines langtech and service in one easy-to-purchase option with a predictable recurring cost (“The Pricing Research Series Reports”).

Global content will become a platform feature. Companies such as Airbnb and Expedia recognize that they are not travel companies, but rather high-tech global content companies with expectations of local-language support at every touchpoint (“Airbnb: A Lesson in How to Implement Language at the Platform Level”).

Similarly, Zendesk sees itself as a customer support solution, regardless of where that customer is. Think of Oracle, SAP, Salesforce, and similar companies as the sinew and muscle of enterprises, scaling to global implementations under any and all languages. They understand the need to make local language available at every point in the processes they manage and have begun integrating translation as a feature into their offerings.

We expect more organizations that are pivoting to D2C to be the first-movers to this mindset. Why? Because, in many cases, their products and services do not function independently of information about them. The most valuable content and code are often generated by third parties. Without this content – delivered in multiple languages – their brands cease to exist.

Pivoting to this mindset requires that language accessibility no longer be relegated as a translation task outside the bounds of product or service development and delivery. Rather, it will be integrated into content and code workflows that affect almost everyone within an organization at some point. That means language is designed, delivered, and optimized as a feature of products, services, processes, and programs from the beginning. As more enterprises adopt continuous localization, their langtech will be considered business-critical, requiring continuity of service, secure translation servers, and immediate fallback solutions (“Continuous Localization at Warp Speed”).

MT developers will ride the metadata wave. Although the hype right now is around translation driven by GPT-3, DeepMind Gopher, and other massive datasets, the next leap forward in machine translation will be found in systems that adapt and respond automatically to context and project information (“Responsive Machine Translation”).

That intelligent reaction will result from pervasive metadata that is purchased from data brokers (including specialist LSPs), systematically harvested from an enterprise’s transactions, and opportunistically gleaned from an assortment of other sources (“Bridging the Multilingual Training Data Divide”). Although the earliest implementations have been laboratory experiments, we predict rapid uptake of responsive features because of the flexibility and quality improvements they can deliver. Expect to see offerings from generalists Google and Microsoft as well as targeted offerings from machine translation specialists. Following on the heels of responsive MT will be responsible machine translation.

Omnichannel marketing will plunge into the metaverse. The online virtual world with augmented reality will need to offer experiences in the preferred language of the visitor. While all the technology components already exist, they will need to be assembled in a new way.

The interactive nature of the space will also put a greater emphasis on on-demand language services versus translations or video localization performed ahead of time. Because the metaverse is meant to tear down physical barriers, it will take more languages to support an inclusive experience for all who visit it. We expect the metaverse to be a lot like omnichannel marketing is today – lots of hope bundled with a lot of dysfunction as the immersive first-generation cyberverse remains an idea in the making.