New Machine Translation Benchmarks and $40M Funding Round for AI Security

Written by Motrjim Power

Published on 26 Jun 2026

Category AI & Translation News

WMT26 targets low-resource languages while Gray Swan secures major funding for testing frontier AI models

Machine Translation Benchmarks Expand to Low-Resource Languages

WMT26 benchmarks have entered their active evaluation phase with new subtitle translation and low-resource language benchmarks. This development represents a significant step in measuring machine translation performance, focusing on languages that suffer from limited training data and linguistic resources, reflecting a growing commitment to improving translation quality for underrepresented language communities in AI technologies.

$40 Million Funding for AI Security Startup

Gray Swan, an AI security startup, has raised $40 million USD in Series A funding. The company helps frontier AI labs including OpenAI and Anthropic "red-team" their latest models to identify vulnerabilities and security gaps before public release.

Industry Awards and Technical Limitations Warnings

Smartling won the 2026 AI Breakthrough Award with enterprise AI translation growing 218% year-over-year, while RWS Language Weaver Pro received the "Machine Translation Solution of the Year" award for the third consecutive year. Meanwhile, experts published warnings about AI's limitations in literary translation and language needs in conflict zones.

Job Opportunities and New Initiatives

The World Bank and European Central Bank announced translator recruitment for their in-house language teams, with openings in Uzbekistan and Germany. Slator also published its annual Language AI 50 Under 50 for 2026, spotlighting fifty startups in the language AI and technology landscape.