.svg){kind=logo}
AI Solutions
Additional Services
Webinar: Lionbridge Samurai™ in Action
AI-powered localization for games that enhances linguist-at-the-core creativity.
Case Study: Multilingual Retail Marketing
New AI Content Creation Solutions for a Sports and Apparel Giant
- RESOURCES
Struggling to meet the demand for multilingual content? You're not alone. For years, translation teams faced tough challenges — tight budgets, limited human resources, and the relentless demand for content. But what if you could deliver high-quality translations faster, at scale, and at a significantly lower cost? With Artificial Intelligence (AI), it’s now possible.
During Lionbridge’s webinar, “Can AI Post-Edit?” experts from Lionbridge and Cisco Systems explored how AI post-editing is transforming translation and localization.
The session, featuring Marcus Casal, Chief Technology Officer of Lionbridge, and Yolanda Cham Yuen, GTS Program Manager at Cisco Systems, addressed the central question: Can AI post-edit using Large Language Models (LLMs) to deliver accurate, reliable, and cost-effective translations at scale? The answer is a resounding yes — with important caveats.
AI can do post-editing, and it can do it very well.
— Marcus Casal, Lionbridge CTO
Want to watch the webinar in its entirety? Use the button below to access the recording.
Automated Post-Editing: How Has It Become the New Reality?
AI post-editing is no longer a distant vision — it’s here, driving real-world transformation.
The Lionbridge Aurora AI™ platform orchestrates the entire global content lifecycle, from ingesting customer data to translating it with Machine Translation (MT), using LLMs for automated post-editing, and then delivering it back into content repositories.
What enables this? A combination of API-driven automation, Integration Platform as a Service (iPaaS), and robust linguistic assets, like Translation Memory (TM), glossaries, and terminology management. By combining MT, LLM-powered post-editing using frontier models, and smart workflows, organizations can deliver content more quickly and predictably than ever before.
But speed and scalability aren’t enough. Quality still matters. That’s why Lionbridge’s approach incorporates Human-in-the-Loop (HITL) oversight. Humans train and tune the models and evaluate outcomes as necessary to ensure the final product meets its content goals.
What Was Cisco Networking Academy’s Practical Application of Automated Post-Editing?
Yolanda shared Cisco’s use of automated post-editing for its global social responsibility initiative, Cisco Networking Academy.
Cisco Networking Academy provides free technology education in networking, cybersecurity, programming, and other data science topics, reaching over 23 million learners in 191 countries. Localization is essential for maximizing the impact of its courses, as language can be a formidable barrier.
To expand access and further make Network Academy courses available to people worldwide, Cisco needed a scalable solution to translate millions of words into more than a dozen languages, often under tight timelines and budgets.
Their answer? The implementation of automated post-editing.
What Did Cisco Network Academy’s Solution Entail?
Cisco’s solution involved the use of:
Translation Memories for previously approved content.
Neural Machine Translation (NMT) for the initial translation due to its speed, cost, and consistency.
LLM-powered AI post-editing to refine the output.
Human testers for functional in-context review — especially for complex languages.
What Type of Results Did Cisco Network Academy Achieve?
The results were nothing short of remarkable.
The use of automated post-editing allowed Cisco Networking Academy to translate over 15 million words into 14 languages, supporting 24 courses in just three months. The bottleneck shifted from translation to functional testing and staging, demonstrating significant efficiency gains. All this work cost less than $70,000, which is dramatically lower than traditional methodologies.
By leveraging LLMs for post-editing, Cisco can now release content in multiple languages simultaneously, significantly reducing the time lag between English and localized versions and expanding global access to its courses.
“We are now seeing content being translated at a speed and cost that we have never seen before. …[Automated post-editing is] opening the door to new spaces, to new scopes that otherwise couldn’t be affordable, couldn’t be feasible in the very challenging scenarios that we have.”
— Yolanda Cham Yuen, Cisco Systems
What are the Limitations and Risks of AI Post-Editing?
AI post-editing is powerful, but not perfect. The speakers highlighted several limitations:
Low-resource languages present greater challenges for both people and machines. LLMs (including frontier models) perform best in English and widely spoken languages; specialized glossaries and training data are essential for niche languages.
Cultural nuance, tone, and domain expertise still pose challenges for AI alone. Using prompt flows and human oversight is crucial for capturing subtle differences — whether in sports terminology or technical language.
Hallucinations — where AI generates inaccurate or misleading information — can occur, sometimes convincingly so. This limitation underscores the importance of functional testing and direct end-user feedback as essential safeguards for high-impact content.
Marcus shared a practical anecdote: The phrase “protect your turf” for a basketball shoe listing on an e-commerce site was translated into Spanish as “césped” (artificial turf or lawn) — suitable for soccer, but not for basketball. By refining glossaries and terminology, the error was corrected, leading to improved user trust.
How Does Human-in-the-Loop Enhance AI Post-Editing?
Will AI post-editing replace traditional translation workflows? Not entirely. Yolanda and Marcus emphasized that while AI opens new possibilities — especially for rapid, large-scale projects — human expertise remains indispensable.
Translators and localization professionals must adapt by:
Developing skills in prompt engineering and workflow automation
Mastering terminology management and brand voice
Partnering with AI to focus on higher-value tasks: tone, audience engagement, and domain-specific content
Providing creative input and quality assurance for outputs generated by MT and LLMs
The industry is shifting toward a “cobotic” model — cooperative robotics — where humans and machines work together to achieve the optimal outcomes.
How Does Integration and Terminology Management Boost Results?
Integration is key. AI post-editing must fit seamlessly with Content Management Systems (CMSs), document management platforms, and other repositories where content lives and evolves. Automated integrations ensure a quick, efficient round-trip for content updates and localization.
Terminology management is equally vital. As Marcus explained, investing in robust glossaries and brand voice assets — especially now that translation and post-editing costs are lower — dramatically improves the acceptability and accuracy of AI-generated translations.
What’s Next for AI Post-Editing in Translation?
Looking ahead, as LLMs continue to improve and more curated training data becomes available, AI post-editing will become increasingly important in localization. However, human creativity, contextual understanding, and ongoing oversight will remain essential for driving innovation and maintaining high translation quality as technology continues to evolve.
What Are the Key Webinar Takeaways?
This webinar provided insights into automated post-editing for enhanced translation workflows, enabling the scaling of global content. Here are the key points:
AI can conduct post-editing well and reduce human effort.
AI post-editing leverages Large Language Models for faster, scalable translation and localization.
The continued use of Machine Translation and effective terminology management (TMs and glossaries) enhances quality and consistency.
Human-in-the-loop involvement remains essential, particularly for specialized content and low-resource languages.
AI-driven service levels — ranging from no human post-editing to the post-editing of all content — may be used to match your content profiles.
Acceptable quality levels may still be achieved with less human involvement.
AI post-editing opens new possibilities for expanding content reach and reducing costs.
Understanding AI Post-Editing: Your Top Questions Answered
We recognize that accuracy has long been a challenge with MT solutions. Lionbridge combines an AI-first approach, using automated quality measurement capabilities, with a human-in-the-loop approach, allowing us to have a pulse on accuracy at all times.
AI is leveraged to enhance the translation output from traditional tools, such as Translation Memories (TMs), Neural Machine Translation (NMT), and RAG.
Our experience with AI post-editing demonstrates that it can meet scalability needs and maintain quality. Still, human oversight is essential to monitor and adjust the tool for accuracy and specific content requirements.
Our AI post-editing solution starts with an initial assessment of the source content to understand its overall context. Editing and validation steps are performed with this context in mind, ensuring that the produced translations align with content goals and/or profiles.
We designed our AI post-editing solution to be configurable — linguistic prompts can be edited and updated based on automated monitoring and human feedback from linguists and customer Subject Matter Experts (SMEs) involved in the translation process.
Additionally, Lionbridge offers other AI solutions that enable us to perform source analysis and produce reports that capture recommended modifications to source content. We can pair these solutions with our AI post-editing solution for further optimization of the content strategy.
Our AI post-editing solution enables us to define and target where human linguistic involvement is most valuable.
Using our REACH framework, we work with our customers to assess content goals and configure the AI solution to optimize translation output. We can then define varying degrees of human involvement, ensuring the level of effort matches the content needs and profiles.
Handling industry-specific terminology that varies by language and region is challenging, but it can be addressed in several ways.
By leveraging metadata, we can provide a broader context for the AI tools and instruct the LLM to address additional requirements, such as regional nuances — assuming the content is properly labeled and tagged. Lionbridge can support companies with data services to meet this prerequisite.
For terminology that must be adapted by municipality or region, we use a RAG framework. In approaching AI post-editing, we establish guidelines for the LLM to perform specific actions based on defined linguistic rules.
The configuration in our solution also allows for referencing external materials as supplemental examples, helping the LLM generate content that is more tailored to specific contexts.
Since this type of content tends to evolve over time, it is critical to maintain and update linguistic prompts. That’s why our AI solutions are built to be controlled and curated with human oversight.
Yes, our AI-first platform solution is LLM-agnostic and not tied to any specific model. While it has been calibrated with OpenAI GPT models, we can work with customers to leverage their own LLM engines instead.
This scenario would fall under a custom configuration and may require additional evaluation/configuration to ensure the LLM meets quality standards.
For these solutions, we recommend collaborating with our Solutions and Language Technology teams to understand needs, goals, and requirements.
Our AI post-editing solution features configurable linguistic prompts that can be adapted to meet changing content needs and regulatory requirements. While we propose an AI-first process, the control and curation of linguistic prompts remain with humans — computational linguists, Subject Matter Experts (SMEs), and language experts — for optimization and calibration.
Generic, untrained, chatbot-based LLM models struggle to address tone, content, and cultural nuance. Our solution addresses this challenge by using a configurable prompt chain approach, which provides the LLM with specific instructions to define style, tone, and terminology, and linguistic guidelines pertaining to the context of the original content. Through controlled configuration of linguistic parameters and prompts, we can leverage AI post-editing to enhance translation processes.
Our solution begins by prompting the LLM to understand the context of the source content. We use that context to guide the LLM’s decision-making and editing, along with instructions for evaluating terminology, editing fuzzy matches, and validating translations at the segment level — ensuring that the AI operates within defined parameters. Our approach also allows for ongoing updates to linguistic prompts, so if hallucinations do occur, we can adjust and correct the issue as needed.
The human testers are typically computational linguists or linguists with experience in prompt engineering. They are responsible for designing, testing, and updating the parameters of our AI post-editing solution. These linguists validate the output and provide feedback, which is then used to adjust and refine linguistic prompts and parameters.
Interested in exploring other AI-related webinar topics Lionbridge has delved into? Visit the Lionbridge webinars page for a library of webinar recordings.
Get in Touch
Ready to capitalize on the power of AI post-editing to meet your translation and localization needs? Lionbridge can assist you in building agile, scalable workflows that combine automation, LLMs, and human expertise. Get in touch with Lionbridge today to transform your content delivery and reach more audiences worldwide — faster, smarter, and with greater impact.
Note: The Lionbridge Content Remix App initially created the recap blog, which a human then refined.
Fill out our contact form to start a conversation with us.
We’re eager to understand your needs and share how our innovative capabilities can empower you to break barriers and expand your global reach. Ready to explore the possibilities? We can’t wait to help.
