Welcome to the LOCANUCU blog, your ultimate destination for localization news you can actually use. In today’s roundup for August 20, 2025, we're covering the industry's dynamic landscape. From major acquisitions like Wolfestone Group's purchase of Omni Interpreting to the release of a groundbreaking AI benchmark dataset by Welocalize and Duke University, there's a lot happening. We'll also explore significant public sector opportunities, new platform integrations from memoQ and Plunet, and the critical conversation around AI's practical application versus its hype.
- UK-based Wolfestone Group has acquired US-based Omni Interpreting.
- The acquisition expands Wolfestone's presence in the United States and adds interpreting services.
- The Scottish Government issued a Request for Proposal for language services valued at £20 million.
- The Scotland contract covers interpreting, translation, and transcription over four years.
- U.S. Citizenship and Immigration Services (USCIS) is seeking providers for remote telephonic interpreting.
- Welocalize and Duke University have launched LangMark, a new open-source dataset.
- LangMark is designed to benchmark human post-editing of AI-generated translations.
- The dataset includes over 400,000 segments in 10 language pairs with quality scores and editing time.
- A recent study highlights the risk of clinically significant errors from AI translation in medical reports.
- The study emphasizes the continued need for human oversight in high-stakes medical translations.
- Smartling has launched a managed container and private cloud (MCP) server for enterprise clients.
- Smartling's new solution offers enhanced security and control over AI translation data.
- Takeaways from the AI4 conference indicate a shift from AI hype to practical ROI and implementation.
- Naitiv founder Gayatri Shahane discussed the importance of finding a specific product-market fit in language AI.
- memoQ and Plunet have announced a new partnership and enhanced platform integration.
- Crowdin provided best practices for translating ambiguous user-generated content.
- A new Master of Science degree in AI Language Technology in New York has gained state government backing.
- The upcoming LEO International Conference has been officially announced for industry professionals.
- GALA has proposed bylaw amendments for 2025, now open for member voting.
- A feature on Tunisia highlights the country as a model for societal multilingualism.
- A BBC report covers the successful community-led revival of a dormant language.
- Industry discussions continue to focus on blending AI efficiency with irreplaceable human insight.
In today's roundup, we begin with significant movements in the mergers and acquisitions space. The UK-based Wolfestone Group has announced its acquisition of Omni Interpreting, a US-based provider. This strategic purchase is set to expand Wolfestone's footprint in the United States, enhancing its service offerings with on-site and video remote interpreting capabilities.
On the public sector front, the Scottish Government has issued a substantial Request for Proposal for a comprehensive language services contract valued at an estimated £20 million over four years. The contract seeks suppliers for interpreting, translation, and transcription services, indicating a strong, ongoing demand for language support in the public sector. In a similar vein, U.S. Citizenship and Immigration Services, or USCIS, is actively seeking contractors to provide remote telephonic interpreter services. This is primarily to support asylum and refugee interviews, a critical function that relies heavily on accurate, accessible language services.
A major development in the field of AI translation research comes from a collaboration between Welocalize and Duke University. The two have launched LangMark, a large-scale, open-source dataset designed specifically for benchmarking the human post-editing of AI-generated translations. The dataset contains over 400,000 translated segments across ten language pairs, each annotated with quality scores and detailed post-editing time. The release of LangMark is a significant contribution, providing a standardized resource for researchers and developers to measure and improve the efficiency of AI translation workflows.
While the tools for AI translation advance, the question of their accuracy in specialized fields remains critical. A recent study examining the accuracy of AI translations for medical reports found that while general-purpose models are improving, they still present risks of clinically significant errors. The findings underscore the non-negotiable need for human oversight and review by subject matter experts when dealing with high-stakes content.
Turning to platform developments, Smartling has enhanced its AI translation offerings. The company announced a new managed container and private cloud server solution, designed to provide enterprise clients with greater security, control, and customization over their AI translation data and workflows. This move caters to the increasing demand for secure, private AI environments in large organizations. Further insights into enterprise AI adoption come from the recent AI4 conference, where a panel discussion shared by Phrase highlighted a shift from hype to practical application. The key takeaway was a focus on achieving tangible return on investment through scalable, real-world AI implementations that balance automation with essential human expertise. On the topic of finding a niche in the AI landscape, Gayatri Shahane, founder of Naitiv, discussed the journey of achieving product-market fit for a language AI startup, emphasizing the importance of solving specific, high-value customer problems.
In the technology partnership sphere, memoQ and Plunet have announced an enhanced integration between their platforms. This collaboration aims to create a more seamless workflow for language service providers by tightly connecting Plunet's business management system with memoQ's translation environment, reducing administrative overhead and improving project efficiency. Addressing a common challenge in localization, Crowdin offered insights on handling ambiguous user-generated content. The platform suggests leveraging context, providing visual aids like screenshots, and using community feedback loops as effective strategies to improve translation accuracy for this type of dynamic content.
Education in our industry is also evolving to meet new demands. In New York, a new Master of Science degree in AI Language Technology has received official backing from the state governor. This program signals a growing academic and governmental recognition of the need for a new generation of talent skilled in both linguistics and artificial intelligence.
For those planning their event calendars, the LEO International Conference has been announced, promising to be a key gathering for industry professionals to discuss emerging trends and technologies. In organizational news, the Globalization and Localization Association, GALA, has put forth proposed amendments to its bylaws for 2025, which are now available for member review and voting.
Broadening our perspective, an article on multilingualism in Tunisia serves as a compelling case study of how Arabic, French, and English coexist and shape its cultural and business landscapes. And in a story that touches the very core of our industry's purpose, the BBC has reported on the successful revival of a dormant language, a powerful reminder of the vital role that language plays in preserving culture and heritage, often aided by modern technology and dedicated communities.
That's a wrap on our August 20, 2025, news brief. Today we covered the strategic expansion of Wolfestone Group, the massive LangMark dataset release from Welocalize and Duke University which promises to standardize MT post-editing research, and the continued push for secure, enterprise-grade AI solutions from companies like Smartling. We also touched on the importance of specialized education, industry partnerships, and the ever-present need for human expertise in an increasingly automated world. For more actionable insights and the latest trends in localization, keep following the LOCANUCU blog.