ULG's Language Services Blog

The Future of Interpretation Integrates AI with a Human Touch

As seen in Globalization and Localization Association (GALA)

Being able to communicate with multicultural, multilingual consumers from around the world has never been more important. Businesses serve a more diverse clientele than ever, and these consumers expect personalized experiences that are smooth, seamless, and technologically advanced.  

Our team of language experts believes that Artificial Intelligence (AI) and its transformative technology will play a pivotal role in creating more personalized, efficient, and culturally sensitive communication solutions. In this article we’ll examine the pros and cons of AI in interpretation and share lessons our team has learned from integrating AI with our services and other technologies such as interactive voice response (IVR) to better serve our customers.  

The Pros of AI in Interpretation 

The hype is real: AI can improve the quality of over-the-phone interpretation services in a number of ways. Here are a few positive changes we’re noticing: 

Efficiency: Serve more customers in less time 

AI has the potential to make interpretation much more efficient. For example, improved call flows can save time for our clients and their customers. They also improve the customer experience, especially if it means they don’t have to wait as long.  

Scalability: Grow your business without adding staff 

AI also offers the potential to scale up the services you offer or the number of customers you serve without increasing the number of staff members. This means providing a level of service that was previously impossible due to cost or personnel constraints. For example, conferences are currently able to provide simultaneous interpretation in many different languages by utilizing AI-powered real-time  interpretation with AI-generated voices, something that would have been cost-prohibitive in the past.  

ULG’s resident MT expert Blanca Vidal described other ways AI can help us grow with our customers in a recent interview with Authority Magazine: “We can create style guides that mimic human language that can connect remote employees, provide healthcare information to multilingual patients or other applications we’re not thinking of yet, all with the touch of a button.”  

Data-driven insights: Utilize large data sets to better serve customers and identify trends 

With AI, we can analyze many different types of data to identify customer needs and trends across cultures. These cultural insights help us to provide more personalized and efficient service, constantly adapting to meet the evolving needs of our clients and their callers. 

Cons of AI Integration 

With all the positive buzz around AI utilization today, it’s vital to acknowledge the limitations of its use cases as well. Here are some reasons our team has found to be cautious:  

Quality: AI is not a substitute for human interpretation  

Efficiency without quality is a recipe for failure, and when cultural nuance and context are critical, AI sometimes falls short. When an AI interpreter squared off against two professional human interpreters, the AI was better able to keep up with the pace of the content than humans.  However, it struggled with word choice and clarity at times. 

The U.S. immigration system has tried to use AI as a substitute for human interpreters. But the cost has been unacceptable: massive human suffering as immigrants struggle to explain asylum claims from detention using AI tools that can’t process their regional dialects.  

Security: AI tools may not be secure  

Publicly available AI tools like ChatGPT feast on massive sets of data for training, including potentially sensitive client details. There’s always the potential for service providers to store, access or misuse those details, or for them to simply be regurgitated in future chat sessions.  

The solution is to use AI tools in a secure environment with the help of a trusted service provider.  

Bias and Ethical Challenges  

We tend to think of bias as a human failing, but AI is not immune to it. These tools adopt human biases from the data they learn from, making them far from impartial. Fixing bias in AI starts with using diverse teams to build these systems and diverse data to train them. Human supervision via post-editing and quality assurance are vital to spot and correct bias in training materials and scripts for interpreters.    

Limited analytical abilities: Why AI's current level of understanding may not be enough 

No matter how intelligent a large language model (LLM) like ChatGPT can sound, these systems don’t think like we do. As a result, LLMs will sometimes “hallucinate,” which means that they will provide plausible-sounding answers that are incorrect.  

There's also the issue of logic and reasoning: While AI can quickly sort through tons of data, it often stumbles on tasks that require complex thinking or deductive reasoning.  

If you're relying on AI for language support, these limitations can be more than just minor hiccups; they can lead to real misunderstandings.  

How we’re using AI at United Language Group 

In translation and interpretation, where the nuances can be as important as the words themselves, our team has learned to blend the best of both worlds: human expertise with AI efficiency. We’re not looking to replace human interpreters who bring valuable skills and expertise that AI can’t replicate. Instead, we’re using AI to improve the customer experience by doing what it does best: analyzing data, spotting trends and powering process improvements.  

AI and IVR to streamline the customer experience 

Our goal is to use AI to help callers get the support they need in their own language as quickly as possible, while freeing up human interpreters to focus on more complex and satisfying work. By integrating AI with our interactive voice response (IVR) system, we can speed up and improve call flows, ensuring that callers with limited English proficiency (LEP) have easy access to a full range of self-help services. This frees up interpreters to focus on what really counts: delivering high-quality, nuanced interpretation that truly bridges language gaps.  

We’re also actively designing ways to route calls to the interpreter who is best suited to handle them. AI tools can instantly connect calls to the most appropriate interpreter, based on past performance, specialization, and even caller preferences, making the whole process smooth and virtually seamless. This isn't just about speed—it's about making each interaction as helpful as possible.  

AI for improved efficiencies 

We’re building a system powered by AI that can use call data to identify routine questions asked by consumers. With this information, we can help our clients create efficiencies, reducing the time it takes to handle these questions, improving the customer experience, and potentially designing proactive solutions that reduce the need for support calls. 

AI to understand the cultural drivers of engagement 

One of the most powerful use cases we’ve found for AI is mapping the Cultural Drivers of Engagement (CDE). These are the factors within a culture that affect how a customer engages with a company, such as their demographics, their belief systems, and the way they research, shop or purchase. With the CDE, we can provide culturally relevant experiences for multicultural consumers that to increase engagement.  

Humans + AI = Success 

AI holds enormous potential to enrich language services, making us faster, smarter, and more in tune with the needs of our customers and the communities they serve. However, AI has strengths and weaknesses.  

We're committed to leveraging the best of both worlds—human expertise and AI capabilities—to deliver secure solutions that drive success for our customers and make the world a more inclusive place. By focusing on creating new, viable workflows we are empowering our clients to communicate with their diverse consumers more efficiently and effectively.  

To see how we can design AI-enhanced solutions for your organization, connect with our interpretation experts today. 

 

James Dodson

James Dodson is a results-oriented executive with 30 plus years of global experience in business strategy, planning, and operations management, Educated at the United States Military Academy at West Point, the Canadian Forces College, and the National War College, James leverages both his business and Army experience to lead the United Language Group’s Interpretation business unit. LinkedIn: https://www.linkedin.com/in/james-dodson-86492645/