Insights from CEOs

What does Post-Localization – and the AI that is driving it – mean for the industry?

Welcome to the mid-year installment of CSA Research's 2024 CEO Insights.

In this 9th edition, we have gathered insights from CEO members of CSA Research's Leadership Council.

*CEO commentary appears in order of company revenue size.

Post-Localization coincides with the dawning of a new era: a truly globalized, multilingual economy. We see it as an incredible opportunity to help our clients dramatically expand their addressable market. Traditionally, translation and localization were approached as the final step in a content journey. This method often limited the potential reach and efficiency of global communication strategies.
With the advent of AI, we can make content multilingual from the outset. AI enables the seamless integration of multiple languages into content creation processes, ensuring that our clients' messages resonate globally from the beginning. This proactive approach revolutionizes how we think about content and its potential impact.

To achieve this, language access must move upstream and become embedded within our clients’ content systems and strategies. Furthermore, our role must evolve to be more consultative, earning a spot in strategic conversations. This means we need to be well-versed in this technology to guide our clients effectively.

Humans remain as integral to this process as they have ever been. They are essential as experts in the loop, heavily involved in program design, implementation, quality control, and optimization. By combining human expertise with AI capabilities, we can deliver unparalleled quality and reach, driving forward a truly global communication strategy.

June 2024

Read January 2024 Statement

Read July 2023 Statement 

 Read January 2023 Statement

 Read July 2022 Statement

Read January 2022 statement 

Read July 2021 statement

Read January 2021 statement

Read 2020 statement

To start, I will challenge the 'exam question' we have been set for this edition of CEO Insights. Whilst I agree with the themes from CSA’s research on 'Post-Localization,' I am not sure that the label it has been given is the best description. The reality is that AI is disrupting our industry, creating new opportunities and challenges for all participants. Rather than sounding its demise, as may be suggested by the term 'Post-Localization,' AI is changing the nature of localization. We see this as more of an 'evolution' of the industry towards more innovative human and machine collaboration.

Anybody claiming to have perfect insight around the impact of AI is probably unwise to do so. All players in our space face the choice of either running towards the opportunities or hiding from the risks. For many years now, we at RWS have believed the right approach is to grab and shape the opportunities and to help our clients to do the same.

AI is changing, not replacing, the role of humans in our industry, and the pace of innovation is exciting. It also creates hazards which we must help our clients to avoid in order to protect their brands. Over 25% of our global revenues are now directly linked to AI, so the transition is tangible and will shape our industry for years to come.

June 2024

Read January 2024 statement 

Read July 2023 Statement 

 Read January 2023 Statement

 Read July 2022 Statement

Read January 2022 statement 

Read July 2021 statement

Read January 2021 statement

Read 2020 statement

As we proceed steadily and quickly into an era of AI-augmented global content solutions, moving away from traditional language-centric structures and strategies, we start to see early results of the efforts, trials, and developments of the past months. While absolute truths are mere marketing illusions, and both technology and mindsets are evolving as we speak, some facts are becoming clear, presenting both challenges and opportunities.
Acolad Labs' recent analysis confirms that while Neural Machine Translation (NMT) still surpasses Large Language Models (LLMs) in quality, performance, and cost-efficiency, the integration of AI into our offerings is revolutionizing the content value chain. On the other hand, the integration of AI into our services and workflows is becoming increasingly tangible. We are witnessing this through the introduction of new offerings, such as AI Voice Over, QA for regulatory compliance, and more. Additionally, AI is enhancing our localization processes with advancements in Automated Post-Editing (APE), Adaptive Quality Evaluation (AQE), and Translation Memory (TM) improvements.
But while it seems promising, it's crucial to discern real outcomes from the AI buzz, requiring close monitoring and evaluation. It is still early to have an exact picture of just what benefits can be drawn from it. It is through progressive and concrete experiments with our customers that we implement the right scenario for them.
Moreover, the industry’s direction will heavily depend on buyers' perspectives and expectations. Will they increasingly push for automated solutions, or will they maintain a focus on high quality with “human at the core” approaches? Regardless of all of this, something is undeniable: Content and Language Services Providers, with their vendors, need to adapt once more, with continuous improvement and “tech addiction.” This time, however, adaptability is not enough: a true global footprint and the resources, partnerships, and capacity to deliver the technology are a must-have.

June 2024

Read the January 2024 Statement

Read the July 2023 Statement

Let’s face it, many language professionals never expected the Post-Localization Era to come so fast. The Translation Bureau has been poised for this change for many years now. Still, as we reflect on the Translation Bureau’s 90 years of existence, we find ourselves relating with the craftspeople whose lives suddenly changed during the Industrial Revolution.
I would say, however, that the change is more complex to navigate today than it was a few hundred years ago—especially when you’re a government agency in an officially bilingual country. Are there truly any low-risk scenarios where AI can handle things by itself when language rights come into play? Where should the bar be placed in ensuring quality when “substantive equality” between official languages is a legal requirement?
Faced with a decline in demand for traditional linguistic services, language professionals must adapt and diversify their offerings, yes. But they must also capitalize on their human prerogative of being the true judges of quality to cut through the cost-saving hype and make sure their clients remain aware of their added value.

  June 2024 

Read January 2024 Statement

Read July 2023 Statement

Read July 2022 Statement

Read January 2022 Statement

Read July 2021 Statement

It’s a fascinating and fast-moving time in our industry as we all explore and apply the capabilities of AI. The pressure to offer solutions that help companies do more for less has always been there, but the speed of change has gone up exponentially. This means that leaders at LSPs need to define and apply new approaches faster than before.

Generally, the opportunity could be huge for us to build new service offerings and approaches that align with Post-Localization. These include:

•    Content generation for highly customized in-market content instead of translating or transcreating content that was created for a home market.
•    Next-level post-editing/quality reviewing. AI-generated translation has output issues that are different from traditional machine translations. For example, AI translations can be full of bias that comes from the data they were trained on. Those errors and omissions need to be found and fixed.
•    Applying AI across the workflow. AI automations can handle things like file prep, subject matter assessments, routing to the right translator, post-translation, quality control, etc.
•    Helping enterprises create, test, and improve their LLMs or GenAI models they have developed for their own products.

And because AI implementation is strategic and revenue-driving rather than transactional, it gives LSPs access to C-suite executives, who are now often demanding that everyone use AI in their area and are keen to have these conversations.

The winners in our industry will be those who automate wherever possible while focusing on dedicating the right human experts to add value wherever needed to help companies go global faster, more cost-effectively, and at optimal quality. The era of Post-Localization demands more of us and will require that the strongest of minds within our industry rise to these challenges.

June 2024

In two words: exciting times. As an industry, we invested heavily over the past few decades to produce ever-increasing amounts of content in more languages. Faster. Cheaper. But the existing language technologies, processes, and to a large extent, our mindsets, have been reaching the limits of what’s possible. AI is changing the game. It’s a massive opportunity. It is a great technology that will allow us to do even more. And even better. And it’s still early days, with a lot more to come. As such, we recognize the necessity of integrating AI into our processes to expand our market reach, enhance efficiency, and maintain high standards of quality.

While AI transforms our workflows, human expertise remains irreplaceable. Our role evolves to more consultative, strategic involvement, guiding clients through technological advancements and ensuring quality control. This human-AI synergy is crucial for maintaining cultural relevance and mitigating the risks of AI.

Our industry must adapt to the rapid pace of AI innovation. This includes developing new service offerings, such as AI-enhanced content generation and advanced quality evaluation tools. It is our mission to educate clients on the effective use of AI, aligning their expectations with realistic outcomes and maintaining the integrity of their brands.

It is also the responsibility of the leaders in our industry to support our supply chain, particularly the linguists who are integral to this process. As AI reshapes workflows and introduces new efficiencies, we must ensure that our linguists are not left behind. This means investing in their continuous education, providing them with the tools and training to work effectively alongside AI, and recognizing the value they bring to our services.

By building a collaborative environment where technology enhances human expertise rather than replaces it, we can navigate this transformation responsibly and ensure sustainable growth for all stakeholders involved.

  June 2024 

Read July 2023 statement

Read January 2023 statement 

Read July 2022 statement 

Read January 2022 statement

Read July 2021 statement

Read January 2021 statement

Read 2020 statement

The Post-Localization Era or AI disruption of the industry. For many, the language services industry seems to be disrupted by AI, whereas it is merely, admittedly, an important phase in the evolution of the language and content management space. I guess a big difference is that content, language, and of course AI are now boardroom as well as general-public topics. Localization or translation of content has historically been handled by a support function within organizations with generally little visibility to lines of business or management. The focus of these outsourced tasks was cost, quality, and capacity. The industry’s response was process automation and optimization.

Still, in the localization era, organizations understood that they could connect better and more with their target audiences (customers, citizens, etc.) by localizing their content. Content grew to be a corporate asset and localization a means to create more value. The industry got exposed to lines of business and became more familiar or integrated within customer business processes or use cases.

So, what does AI change? It does indeed allow us to further automate tasks and thus drive down costs. Most importantly, it enables us to do more with content and to generate or transform content in many ways. Managing content still won’t be everyone’s core business, and organizations continue to need the help of technology and service providers. The industry’s historic revenue drivers, content volume combined with unit of work price, are, of course, impacted. A trend where volume growth gets offset by lower prices is not new but accelerating.

We can nevertheless expect the content and language industry players to transform their business models and content processing platforms to be even more robust, integrating all available AI/GenAI tools as they become available. The added value lies in the orchestration of fit-for-purpose processes to deliver the right content at the right time and in the appropriate format. This orchestration includes technology and human intervention, which is required to secure proper content governance.

In conclusion, the industry has the DNA, resources, and competence to meet the challenges of content management in the Post-Localization Era. We will continue to leverage human expertise and technology specializing in customer-centric use cases.

June 2024

Read January 2024 Statement

Read July 2023 statement

Read January 2023 statement  

Read July 2022 statement  

Read January 2022 starement 

Read July 2021 statement

Read January 2021 statement

Whether or not we are in a Post-Localization Era can be the subject of some debate, as a vast number of organizations are still trying to figure out how to trust AI, let alone harness and leverage it. This does not negate the fact that AI will inevitably drive the transition from language-centric tasks to global content creation: the possibility to quickly create multilingual content—albeit generic and unimaginative—at little cost is too commercially attractive for this trend to be reversed.

From a purely translation standpoint, what we are most clearly noticing—at least within the most discerning organizations—is the adoption of distinct quality levels, depending on the nature and importance of the content to be localized. Interestingly, the “responsibility” for the quality of the localized content seems to be shifting from the LSP to the user, who is now tasked with choosing how their content will be translated or adapted, whether it is using AI, content-curated machine translation, MT with varying levels of post-editing, or professional translation—a process that now incorporates all of these elements.

Until the next real paradigm shift, we believe that the best LSPs will navigate these changes by offering comprehensive, value-added AI solutions combined with specialized professional services, empowering clients to make informed choices about their content's translation and adaptation.

  June 2024

Historically, content has been created by humans, localized by humans, and then digested by humans. For several decades, various technologies have facilitated the localization process, such as Computer Assisted Translation Tools, Translation Management Systems, as well as Machine Translation. However, in the last few years, new technologies on the market have disrupted not only the process itself but the entire content workflow, both spoken and written, creating uncertainty and posing a fundamental question on the future of the industry as it is perceived today. Displacement of human-only services by machines is no longer the imaginings of science fiction but a plausible, near-future state.
At TRSB, however, we firmly believe that the future is bright, and, over the course of the next few years, the translation industry will need to evolve and officially take ownership of what it has always been: an industry of content experts. Customers who have been befuddled by the media hype surrounding Artificial Intelligence will need these experts to help see through the fog in order to appropriately curate content, safely store and move content, apply the most appropriate tone and accurate terms to both serve and reach a broad audience of markets and market segments, and facilitate the use of an ever-growing number of complex technologies in order to achieve their business goals. The notion that technology will eliminate the need for human expertise is a fallacy, insofar as human communication is concerned. Humans need and will continue to need technology to perform various tasks—but humans need to connect with other humans at the most fundamental level. If the translation industry adopts a broader approach to language and content, and we finally acknowledge that our collective inferiority complex has been holding us back all these years from clearly demonstrating to the world that our vast and deep knowledge not only has value but is crucial in serving our customers, then we will continue to thrive in an ever-changing world. With or without AI.

June 2024

Our industry has undergone major changes over the past year. Technology, and AI in particular, has had a major impact on the development of the translation market.

Fears that AI would replace the work of linguists entirely and leave most LSPs without any business were greatly exaggerated. As with MT, which came to the industry 25 years ago, the AI hype is fading. But AI is finding its place in the toolkits of leading LSPs and clients. We must not forget that human beings are at the center of every business, and the translation industry is no exception. While technology is important, the role of professional linguists in the industry has never been greater.

I am pleased that our company has found a successful combination of technological development and expansion of human resources potential. This is essential for future stability and is a guarantee of quality services for our clients.
 

June 2024

Read January 2024 Statement 

Read July 2023 Statement

Read January 2023 Statement 

Read July 2022 Statement 

Read January 2022 Statement

Read July 2021 Statement

Read January 2021 Statement

Read 2020 Statement

Language services are thriving, reflecting a promising outlook for our industry. As we see the adoption of technology at an incredible pace, the need for humans at the core continues to grow. This is a time of innovation and reinvention, where new content creation is increasingly multilingual and language access continues to expand. However, it's essential to remember that AI is a tool that still requires adult supervision.

Government policies and regulations are starting to shape our landscape. In the US, updated healthcare regulations (Section 1557) now explicitly prohibit discrimination and bias in AI-driven translations, mandating human review for critical translations. The EU’s Artificial Intelligence Act categorizes AI according to risk, underpinning the value of human validation.

Leaders in our field have been diligent in lighting the path forward. Guidance on the safe and ethical use of AI in interpreting, developed by the SAFE-AI Task Force, has recently completed a round of public review and comment and is due for publication in the upcoming days. Additionally, the ASTM is incorporating translation labels, reinforcing the importance of transparency and quality assurance for AI-driven translation.

Despite the rapid technological advancements, three billion speakers of marginalized languages are still on the wrong side of the digital divide. However, emerging technologies provide great hope, potentially enabling these communities to leapfrog into the digital era, accessing education, services, and opportunities previously unavailable in their languages.

Our commitment remains to harness AI responsibly, and with thoughtful integration, we can shape a more connected and equitable world.

June 2024

Read January 2024 Statement

Read July 2023 Statement

Read January 2023 Statement 

Read July 2022 Statement 

Read January 2022 Statement

Read July 2021 Statement

Read January 2021 Statement

Read 2020 Statement

The first half of 2024 was definitely characterized by an increased demand for the usage of machine translation and AI-influenced workflows combined with consulting services on the topic of AI. In addition, macroeconomic circumstances mean that companies have to economize on many projects and are therefore looking for ways to reduce costs. The continuing demand for digital processes, improvements in productivity, and the associated fall in costs are fueling the use of modern technologies.

Our customers definitely expect us—as experts in language management—to advise them on these topics and suggest the best solutions, including the corresponding workflows. The problem here is that many users still have no real idea of how to use them correctly, and expectations and reality often diverge.

In our eyes, we have to find the best possible solutions for the customer and user of the translation environment, which is certainly part of Post-Localization; offering a simple translation or localization is no longer nearly enough, as they really need process consulting.

The price pressure is increasing, and customers want to pay less money for the same service due to supposed AI improvements. Now we need to convince the customer that AI will give them more service for the same price, which they can then use profitably. I am excited about future developments and where the journey will take us.

  June 2024

There is an awful lot of change at the moment, and that can be unsettling. Along with writers, clients and agencies are having to evolve very quickly, separating the AI hype from reality, and often finding their way in the dark. I am very optimistic about the future, however, and believe there will be huge opportunities as our industry is able to take its influence further.

With ever more powerful tech at our disposal, I expect that, as ever, it is human talent that will make the difference. The agencies that thrive will be those who are able to best support their clients by orchestrating a network of different platforms and processes that are truly tailored to their requirements. Over the past six months, we have also seen greater awareness of what gen AI cannot do, and where there really is no replacement for good creative writing. The best linguists will be ever more in demand as a differentiator.

In light of this, it is clear that commercial models will need to change. Per-word or time-based structures do not reflect the shifting roles and structures within contemporary localization setups. This is not in anyone’s interests, whether buyer, vendor, or freelancer. I would love to see more outcomes-based pricing, where agencies and clients’ interests and incentives are aligned, and great work is rewarded accordingly. Bringing about this sort of change will take a concerted effort across the industry, involving all stakeholders. If we can get it right, however, it will set everyone up for success in the new era.

  June 2024

Read January 2024 Statement

Read July 2023 Statement

Just like with all disruptive technology in our industry, Post-Localization, driven by AI, means an evolution. While AI is helping us rethink our strategies and workflows, it doesn't replace the need for human expertise—in fact, it elevates it.

The true potential of AI in localization is achieved when it's combined with human expertise. Expert linguists, subject matter experts, and skilled project managers are crucial for managing automations and integrations effectively.

The rise of AI in localization also means an increase in the volume and diversity of global content. Machine augmentation drives growth, allowing us to reach broader audiences more efficiently. Yet, AI alone cannot guarantee quality or cultural sensitivity. This is where human input remains indispensable.

I'm a strong believer in the "humans at the core" concept. Utilizing AI responsibly and ethically is key. AI can enable expansion and streamline processes, but it cannot function independently. The value of human experts—those who bring cultural insight, creativity, and critical thinking—will only increase as AI becomes more integrated into our workflows.

June 2024 

Read January 2024 Statement

Read September 2022 Statement

The concept of Post-Localization is emerging as a significant change, propelled by advances in artificial intelligence. This evolution is drastically transforming our approach to international content. For us as language service providers, this paradigm shift presents both challenges and opportunities. By moving beyond traditional localization and integrating AI-driven processes, we can maintain our growth and become more relevant. With more comprehensive and personalized services, multilingual content delivery, customer experience, and global business growth are enhanced and optimized.

However, in this new era, the role of human expertise remains crucial. AI-powered tools can improve efficiency and accuracy, but the ability to craft creative content that resonates across diverse audiences and the ability to understand context, cultural subtleties, and language nuances such as puns, slang, or expressions are irreplaceable human contributions. As companies continue to expand into new markets, the demand for accurate, culturally relevant, and strategically appropriate content will only increase. It is therefore imperative for us to adopt a forward-thinking mindset, leveraging both AI and human expertise to navigate the complexities of the Post-Localization landscape.

Our role in this transformation is clear: to combine advances in AI with our deep understanding of cultural contexts to ensure our customers receive the best possible service. By embracing the Post-Localization Era, we are positioning ourselves for continued success in an increasingly globalized world.

June 2024

Read January 2024 Statement 

Read July 2023 Statement

Read July 2022 Statement

Read January 2022 Statement

Read July 2021 Statement

Read January 2021 Statement

Since Generative AI went public over a year and a half ago, we can now begin to discern between the true disruptive impact of this technology and the effects inflated by marketing hype. The widespread marketing campaigns associated with Generative AI have raised expectations considerably, leading many decision-makers to pursue AI-driven solutions indiscriminately to optimize processes, reduce costs, and save time. This often occurs without specifying whether the AI in question is generative or another type, resulting in potential misapplications and unmet expectations.

In translation services, distinguishing between Generative AI and Neural Machine Translation (NMT) AI is essential for strategic differentiation. NMT AI, specifically designed for translation tasks, often outperforms Generative AI in terms of accuracy and fluency, particularly for language pairs that do not include English. This strategic distinction is not merely academic but key for ensuring high-quality translations.

Human oversight remains vital not only for editing or post-editing AI-translated or AI-generated content but also for validating the accuracy and appropriateness of outputs. Keeping human translators and editors at the core brings context, cultural understanding, and nuance that AI currently lacks. This combined approach can significantly enhance translation quality and reliability.

The disruptive potential of Large Language Models (LLMs) and Generative AI extends beyond translation and localization into areas such as content creation, text adaptation, quality assurance, and the development of interactive bots. These applications demonstrate AI’s versatility and its ability to augment human creativity and efficiency.

As Generative AI becomes more prevalent, it raises important ethical and legal questions. Recent legislation and ethical guidelines emphasize the need for transparency, accountability, and respect for intellectual property. Internet search engines may also penalize AI-generated content, affecting its visibility and reach. This underscores the importance of integrating ethical considerations and compliance into AI deployment strategies.

Looking to the future, the deployment of Generative AI holds incredible potential, but it must be approached with a balanced perspective. By distinguishing between different types of AI and maintaining human oversight, we can leverage AI’s capabilities effectively and ethically. Staying informed about technological advancements, legislative changes, and ethical considerations will be crucial in harnessing AI’s full potential while mitigating its risks.

  July 2022

Read January 2024 Statement

Read July 2023 Statement

Read January 2022 Statement

Read July 2021 Statement​

Read January 2022 Statement

In a word… rethink. It’s time for the industry to rethink the entire localization lifecycle—from content creation through the publishing of multilingual assets. And it’s time to rethink the value equation—who gets paid, how, and how much.
While some focus on AI as a means to machine translate content, there are myriad other opportunities such as source content assessment, glossary creation, TM maintenance, transcreation, quality estimation, task assignment, and even efficiency reporting. As an industry, we should focus on where our expertise adds the most value and not be concerned with AI replacing more mundane tasks. This leads to reimagining the economics of the industry by putting in place a utility-based system where end users pay based on the value of the translated content versus simply by word. Here, AI is used to unclog the backlog of content to be translated by performing more tasks in the pipeline for lower utility content, freeing up higher-value (and higher-cost) work for humans. I am bullish that the more we use AI as an industry, the more revenue we will generate… for both technology and human-led services.

  June 2024

Read January 2024 Statement

The language services industry is experiencing an undeniable rebirth. AI has opened the floodgates to language automation, and Language Service Providers (LSPs) and tool developers alike are feverishly pushing new horizons of what’s possible. The relentless pace of AI innovation, marked by rapid development of AI tools, product announcements, and extraordinary performance claims, is dizzying. Keeping up with these advancements has become a full-time occupation for tech innovation teams within LSPs. Although concerns about regulation, security, and quality keep some buyers cautious, many have now wholeheartedly embraced the speed, cost-effectiveness, and quality improvements AI offers.

However, navigating this AI-driven landscape is complex. The profusion of machine translation (MT) engines, advanced large language models (LLMs), and a surge of AI tools and applications present a significant challenge for buyers. They increasingly depend on their trusted LSPs for reliable, impartial, and intelligent guidance through this new and vast AI maze. The performance gap between traditional localization, machine translation, and LLMs is rapidly closing, often shifting positions. Neural machine translation (NMT) and LLMs are in fierce competition for dominance. Consequently, LSPs are now able to offer tiered service levels that are both fit-for-purpose and budget-friendly. In the realm of digital media content, AI-driven solutions such as synthetic voices and automated subtitling are no longer just alternatives but often the preferred choice.

Despite the enthusiasm for AI, the industry faces significant challenges. Many LSPs must restructure their operations and service offerings to adapt to the new AI reality, addressing revenue erosion from traditional services. Some buyers remain hesitant, pausing their investments, and those who have embraced AI are not yet fully reinvesting their savings into new global content initiatives. Ultimately, those LSPs that demonstrate agility, innovation, and resilience to ride out the current Post-Localization blues will be well-positioned to capitalize on the forthcoming global content explosion that AI is set to trigger.

  June 2024