There is nothing quite as exhilarating in product development as presenting what you have built to the world. At this virtual event, we didn’t just showcase a collection of new tools and features—we systematically presented a new way for enterprises to operate without borders in the AI era. Here are the five most significant announcements and what they mean for global organizations.
1. We Have Launched an Entirely New Way of Working with Language
DeepL has now become the unified language operating layer for the enterprise. We have built a multilingual AI platform that fundamentally changes how organizations work with language. The new DeepL Translator platform delivers a more efficient language workflow—an alternative to both traditional TMS and LSP models that combines speed, controllability, and transparency with superior translation quality. Teams can generate accurate, compliant, brand-aligned translations directly within any tool or system they use, completing the full translation process in minutes rather than weeks.
This is not an incremental improvement to an existing workflow. It is a different workflow entirely. By eliminating the extract-translate-reinsert cycle, Translation Flow removes the friction that has made localization a bottleneck for content-dependent teams across the enterprise. The platform handles file formats natively, enforces terminology and style rules automatically, and routes content to human reviewers only when quality assessment indicates it is necessary. The result is not just faster translation—it is translation that happens without anyone thinking about it as a separate activity at all.
2. Real-Time Voice-to-Voice Translation for Every Type of Business Communication
Voice-to-Voice translation is now live, enabling real-time semantic transfer across languages. It delivers immediate understanding, more natural conversational flow, and a fundamentally more inclusive communication experience. Whether it’s virtual meetings, in-person meetings and group discussions, customer support calls, or other critical communication scenarios, this capability performs reliably at scale.
The implications extend far beyond meeting convenience. When language barriers dissolve in real time, the composition of global teams changes. Organizations can promote talent based on expertise rather than language fluency. Customer support can be routed to the most qualified agent, not the one who speaks the customer’s language. Safety briefings reach every worker simultaneously in their native language. This is not a convenience feature—it is a structural change in how global organizations allocate human capital.
3. AI Designed Specifically for Language Professionals
Translation Flow pulls data directly from the content systems where it lives and orchestrates the translation workflow end-to-end. DeepL Quality Assessment brings human review into the process precisely where it is needed, rather than applying it indiscriminately. The professional editing experience within Translation Flow guides reviewers directly to the segments that need their attention. The Customization Hub scales linguists’ knowledge and decisions across the entire translation process, transforming individual expertise into systemic quality.
This is a critical distinction from the narrative that AI replaces human linguists. DeepL’s approach is the opposite: AI handles the mechanical translation work at scale, freeing linguists to focus on the high-value decisions where their expertise adds the most impact. Style rules, glossary entries, and quality assessments become force multipliers for linguistic expertise rather than substitutes for it. The linguist’s role evolves from translator to quality architect—designing the rules and standards that guide the AI rather than performing every translation manually.
4. Superior AI Translation Quality That Language Experts Prefer
This month, Slator published a market assessment of AI captions in real-time voice translation, and DeepL Voice significantly led all competitors in both translation quality and stability. 96% of human experts preferred DeepL Voice’s translation output. In the most recent round of blind testing, DeepL’s new model won 94% of all comparisons: 100% win rate against Google Translate, 100% against ChatGPT-5.2, 100% against Microsoft Translate, 88% against Google Gemini 3 Pro, and 81% against Claude Opus-4.6. These results span 16 high-priority language pairs and 48,000 individual blind comparisons conducted by professional linguists.
The most significant finding is not just that DeepL wins, but that the margin of victory widens on the dimensions that matter most for enterprise use: consistency across document types, handling of industry-specific terminology, and preservation of tone and register. These are the dimensions where general-purpose LLMs, trained primarily on internet text, struggle most—and where purpose-built language AI has the clearest advantage.
5. Embedding Instant Translation Through the DeepL API
We are opening our leading capabilities to developers through the DeepL API and DeepL Voice API. This enables developers to embed advanced language AI capabilities into their own products, platforms, and workflows. Whether it’s Articulate helping content creators generate multilingual course materials, or Amazon Connect integrating real-time voice translation into customer support platforms, our APIs are helping enterprises convert language capability into a growth driver.
Nearly 70% of businesses in the United States alone report that language barriers cause daily business disruptions, and 60% say these barriers have slowed their global expansion. The capabilities we have demonstrated are positioned to change exactly that. They form a new growth engine that transforms language from a barrier into a business advantage—not through incremental optimization of old workflows, but through a fundamental reimagining of how language operations can work when AI is native to the architecture rather than bolted on.