NEW YORK, NY – 24/05/2026 – (SeaPRwire) – As global enterprises intensify efforts to engage customers across multiple languages and markets, demand is growing for AI translation platforms capable of balancing speed, accuracy, automation, and governance. Responding to this shift, Smartling announced a broad new suite of AI-powered innovations aimed at helping organizations scale multilingual content operations with greater efficiency and confidence.
The latest release represents the company’s largest AI-focused product expansion to date and introduces several new capabilities spanning translation quality evaluation, large language model optimization, automated localization workflows, and multilingual visual content adaptation. Smartling said the enhancements are designed to help enterprises streamline global communication strategies while maintaining brand consistency and operational oversight.
At the center of the announcement is LQA Agent, an automated translation quality evaluation system built around the industry-standard Multidimensional Quality Metrics (MQM) framework. The solution identifies translation errors, categorizes issues based on severity and type, and generates quality scores and pass/fail ratings across multiple languages and translation methods. The company stated that enterprises can deploy the system in fully automated workflows or integrate human review processes for flagged content before publication.
According to Smartling, internal testing demonstrated that LQA Agent achieved a 90 percent agreement rate with human reviewers and delivered 99 percent accuracy when identifying severe translation errors. The capability is intended to help organizations improve scalability while reducing manual review workloads.
Smartling also introduced Auto Select LLM, an enhancement to its existing Auto Select technology that automatically determines the most suitable large language model for translation tasks. The system incorporates prompt engineering, continuous benchmarking, and retrieval-augmented generation (RAG), enabling AI models to reference customer-specific translation memories and glossaries in real time. The company said the feature allows enterprises to improve translation quality without altering existing workflows.
Another newly introduced capability, Style Rules for AI, applies regional language preferences, formatting conventions, punctuation standards, and custom brand guidelines directly to AI-generated translations. Smartling noted that default configurations currently support more than 30 locales, while additional custom rules can be developed to align with enterprise branding and localization requirements.
To support real-time multilingual communication, Smartling launched Instant AI Translation, a synchronous API-based translation capability designed for integration across enterprise content ecosystems. The functionality supports platforms and creative tools including Adobe Experience Manager (AEM), Drupal, WordPress, Photoshop, Illustrator, InDesign, and Figma, enabling organizations to generate AI-assisted translations instantly within existing workflows.
The company additionally announced Language Adaptation, a feature that enables enterprises to localize existing translations into regional language variants without requiring complete retranslation. Examples include adapting US English into UK English or Brazilian Portuguese into European Portuguese. Powered by large language models, the solution is intended to improve localization quality while accelerating multilingual publishing timelines.
Expanding beyond text-based localization, Smartling previewed AI Image Translation, a technology capable of extracting text from images, translating it, and reconstructing the original visual asset with translated content embedded in place. The company said the capability is intended to help organizations localize banners, product imagery, social media assets, and other visual materials more efficiently while preserving visual consistency across markets.
Smartling stated that responsible AI governance remains central to its product development strategy. The company recently achieved ISO/IEC 42001:2023 certification, an international standard for AI management systems. Combined with ISO 27001, HITRUST e1, SOC 2, and HIPAA compliance programs, Smartling said it maintains a comprehensive framework focused on AI security, trust, and regulatory compliance.
The company confirmed that LQA Agent, Auto Select LLM, and Language Adaptation became available beginning May 20, 2026, while AI Image Translation, Style Rules for AI, and Instant AI Translation are currently being introduced in preview form.
The newly announced technologies are scheduled to debut during Smartling’s Global Ready Conference on May 20, the company’s annual event dedicated to AI translation innovation and multilingual content operations. The conference is expected to feature speakers and enterprise leaders from organizations including Spotify, IHG Hotels & Resorts, DocuSign, DoorDash, and SumUp.
About Smartling
Smartling is an AI-powered translation and localization technology company focused on helping enterprises create, manage, and scale multilingual content operations. Its LanguageAI™ platform combines AI translation, workflow automation, and quality assurance capabilities to support enterprise-grade multilingual communication across global markets.