AI versus Language Service Providers: What does the future hold for LSPs?

March 5, 2024
Post feature
Will AI mean the end of Localization Service Providers (LSPs)?
As is the case in many content-centric industries, this question is front of mind for many in the localization industry. Artificial Intelligence (AI) is now widespread and the number of potential use cases grows daily. So, is it possible to remove the human touch completely?
Lessons from machine translation
Since the 1950s, AI has been used to translate content from one language to another. In the last decade the use of machine translation (MT) has mushroomed. Many in the localization industry (and beyond) prophesized that MT would render human translators obsolete.
In reality, this hasn’t happened. AI has not replaced human translators. Instead MT is a complementary service. Some content and use cases are well-suited to MT, others are best suited to human translation.
The widespread implementation of MT has also taught the industry a vital lesson about the need for human expertise. As is the case with all tools and technology, successful implementation requires expert set-up and continual optimization.
MT is driven by data, and that data must be orchestrated in order to provide high quality outputs: What pieces of data are relevant? What level of quality is appropriate? In short, what data exists and how should we use it? If this optimization does not take place this can result in Data Science Lesson 101: Garbage In, Garbage Out.
What about LSPs?
So how does this relate to LSPs? Just as it was feared MT would be the death knell for translators, Generative AI (Gen-AI) is now being touted as the end for LSPs and localization professionals more generally.
If you can use Gen-AI to create content in the required language, do you need people to oversee and coordinate localization efforts?
At Rubric we believe that both Gen-AI and AI tools as a whole will augment the role of localization professionals, changing their role. Their focus will shift away from translation. Instead they will play a vital role in orchestration, oversight and optimization of both the data which drives AI tools (Large Language Models and beyond), and the tools themselves.
Leading industry body CSA Research recently summarized this well, forecasting that successful LSPs will become:
"…the glue that integrates the people, processes, and technologies that combine to produce winning global customer experiences"
CSA Research, The Evolution of Language Services and Technology, November 2023, p.24
This message was echoed when Rodrigo Fuentes Corradi predicted that LSPs must act as:
"the custodians of the key components of technology, talents, and processes when it comes to language when data is also in the mix"
Rodrigo Fuentes Corradi, What Lies Beyond the Language Barrier for Language Service Providers? MultiLingual Magazine, January 2024, p.44
Rubric’s approach
At Rubric our tech suite is constantly evolving to serve the needs of our clients. We pride ourselves on making best use of technology whilst ensuring our services are "human at the core". That means partnering with our clients to help them navigate the ever-changing tech landscape — developing tools, ecosystems and roadmaps appropriate for their industry, brand and data set. In short, orchestrating their global content efforts, using AI as part of our toolbox.
AI is an important part of our offering at Rubric. But using AI successfully requires a deep understanding of the suitability, ROI, capabilities, and limitations of such tools. As Rodrigo concludes:
"While creating new steps in the localization process, each of these component technologies will also need best practices and governance in terms of quality, transparency, ROI, and security."
Rodrigo Fuentes Corradi, What Lies Beyond the Language Barrier for Language Service Providers? MultiLingual Magazine, January 2024, p.41
A cautionary tale
In our recent roundtable discussion on AI in localization, Arle Lommel from CSA Research provided a cautionary tale about companies that had jettisoned their localization teams, believing they were no longer required:
"Earlier this year, we became aware of a couple of companies in Silicon Valley that let go of their localization teams because they thought 'Gen AI can just replace them.' They quickly found out that these teams were doing a lot more than translation. Now they have let go of a decade's worth of institutional knowledge that they're now going to have to rebuild from scratch."
The invaluable human element: What AI can't do
What AI cannot do is offer the seasoned eye of a professional language expert. It cannot apply reasoning, critical thinking, and decision making. All of which are vital in order to get the best out of the rapidly evolving tech landscape.
One prevalent theme that emerged during our roundtable discussion was how much language service providers (LSPs) do besides ‘just’ translation. Some examples are:
  • Providing cultural understanding in contexts where there is limited data.
  • Handling new terminology and concepts that were not present in the AI tool's training data.
  • Ensuring that a system or process meets all of a company's requirements for security, reliability and ROI.
  • Incorporating institutional knowledge and making decisions based on the specific needs of a company.
  • Providing higher level input like data optimization, product design, content strategy, and project or workflow supervision.
Despite some people's fears, AI is very unlikely to replace the role of localization professionals entirely. However, one thing is certain… it is changing the nature of some people's roles.
Arle said:
"An analogy that I heard recently, that I really liked, is that we're at a point where language professionals have to shift from being cooks to being chefs. They are not just trying to cook up a 'tasty dish of words.' They have to be using all the technology and running the kitchen."
This new approach requires a mindset shift. It means that we all have to work out the best ways to use the technology instead of "holding onto what's been in the past."
How to make the most of AI
If AI doesn't remove the need for localization professionals, how can you make the most of this technology?
At Rubric, we have been helping companies incorporate AI tools into their workflows for years. The key is to focus on the overall process rather than simply looking at the promise of a single AI tool.
During our roundtable discussion, Dominic Spurling, Rubric's IT Director, explained:
"We have always been focused on process optimization, having very robust platforms, and managing content. We continue to invest in this sort of workflow optimization."
As the technologies continue to mature, we are ready to plug them in at any stage of the localization pipeline. Many stages of a content localization process can benefit from AI, from authoring right through to release and publishing. The key is to lean on the knowledge of experts who can help you identify which AI technologies will be most beneficial for your specific workflow and orchestrate your global content efforts.
If you'd like to learn more about AI in localization, make sure to watch our roundtable discussion or read our AI eBook