In this newsletter, we explore how AI is advancing enterprise localization through multiple lenses:
- Delivering gender-inclusive translations with AI agents
- Understanding new MCP and A2A protocols that connect AI systems and enable seamless collaboration across departments
- Learning from Amadeus how AI creates new skilled positions and allows humans to focus on higher-value work
- Discovering how a global Food & Beverage leader created authentic multilingual experiences directly in ServiceNow
Deliver gender-inclusive translations across languages with AI Agents
Traditional MT struggles with gender inclusivity when translating from English to gendered languages. Read how AI agents deliver both grammatical correctness AND inclusivity.
Unlike glossaries, pattern matching, and manual post-editing, AI agents deliver accurate, context-aware translations that scale easily across languages with minimal human oversight.
AI agents enable communication that connects with diverse audiences in their native languages, respecting their linguistic norms and identities.
How to understand the difference between API, MCP, and A2A in Enterprise Localization
APIs are not enough anymore. Now, MCP (Model Context Protocol) and A2A (Agent-to-Agent Protocol)—two new protocols—help AI systems access tools, use business context, and collaborate across functions.
MCP (proposed by Anthropic, with support from OpenAI and Google) and A2A (introduced by Google) protocols transform how AI systems work together.
MCP connects LLMs with external systems without hardcoding every detail, while A2A enables multiple AI agents to collaborate on complex workflows.
Learn from our blog post how these protocols transform enterprise localization and enable Translation AI Agents to serve every department, from customer support to legal, providing consistent translations throughout the organization.
The brave new world of AI in Enterprise Localization: Amadeus’ journey
Many fear AI will replace human jobs with the press of a button. But what if AI isn’t eliminating jobs but transforming them, creating new skilled positions, and enabling humans to focus on higher-value work?
Read a blog post on how our client Amadeus, one of the world’s largest travel technology companies, reimagined their entire approach to language and localization with AI. When one editor couldn’t handle all content from non-native English speakers, they analyzed which language issues appeared repeatedly and used LLMs to fix the “magic 70%” that could be fixed automatically.
Working with Intento and MemoQ, they developed two key processes: Source Quality Improvement (SQI) using LLMs to enhance English source texts and Automatic Post-Editing (APE) applying similar technology to improve machine-translated content.
The review time has been reduced by 60-70% across all languages with experts now focusing on what customers actually care about—correct industry terminology—instead of perfect grammar.
How a global Food & Beverage Enterprise made ServiceNow® multilingual
Explore how one of our customers enabled teams worldwide to access support in their native languages while maintaining efficient and centralized content management—directly in ServiceNow.
With Intento Language Hub for ServiceNow, they created an authentic language experience where employees could access the service portal, submit requests, find information, and chat with support teams in any language. Our customer transformed their entire ServiceNow environment—the Employee Center, Service Portal, Incident processes, and Chat functionality.
New image formats available in the Intento Translation Portal
We’ve expanded our supported file types to include jpg, png, and bmp image formats. Now, you can translate even more content types directly in our Translation Portal, including images containing text.
Combined with our existing support for txt, html, htm, xml, csv, docx, pptx, pdf, xlsx, srt, xliff, tmx, and csv files—plus OCR technology that converts scanned documents into translatable text—you can handle any content format in one place.
The new languages from DeepL and Microsoft are now available in the Intento Language Hub
DeepL: he (Hebrew), th (Thai), vi (Vietnamese)
Microsoft: be (Belorussian), lb (Luxembourgish)
New models and instances for Anthropic added to the Intento Language Hub
New Anthropic models:
- claude-opus-4-20250514
- claude-opus-4-0
- claude-sonnet-4-20250514
- claude-sonnet-4-0
New Anthropic instances:
- ai.text.translate.anthropic.claude-opus-4-0
- ai.text.translate.anthropic.claude-sonnet-4-0
New Google models (Anthropic):
- claude-opus-4@20250514
- claude-sonnet-4@20250514
New Google instances:
- ai.text.translate.google.vertex.anthropic.claude-opus-4-20250514
- ai.text.translate.google.vertex.anthropic.claude-sonnet-4-20250514
You can access all instances through the Intento Language Hub.
Keeping you updated on the latest AI developments in the market
- Anthropic launched Claude 4 with enhanced reasoning capabilities
- Google released new Gemini 2.5 models: Two Pro versions (gemini-2.5-pro-preview-06-05 and gemini-2.5-pro-preview-05-06) and one Flash version (gemini-2.5-flash-preview-05-20)
- DeepSeek introduced updated R1: Latest version DeepSeek-R1-0528 now available
- Bond Capital published a comprehensive “Trends – Artificial Intelligence, 2025” report
- Mistral AI debuted Magistral, their first dedicated reasoning model
- OpenAI has lowered the price of o3 by 80% compared to its previous version, and has added o3-pro, a model that uses more compute to think more deeply and consistently provide better answers
