Despite surging enterprise AI budgets, a surprising number of businesses remain anchored to manual translation. A new report from DeepL on the state of Language AI in 2026 reveals a fundamental disconnect: organizations are pouring capital into AI without redesigning the very workflows that are supposed to benefit from it.
The data paints a stark picture of this inertia. A full 35% of global businesses still rely on entirely manual translation processes. Another 33% use traditional automation that requires human review. This means a mere 17% have adopted next-generation AI, such as LLMs or agentic systems, for translation. The result is that a staggering 83% of companies are operating with outdated infrastructure, which explains precisely why so many AI investments fail to deliver their promised productivity gains.
Pressure to abandon these legacy systems is mounting. Companies are chasing clear benefits from AI-driven translation, citing improved customer experience, higher employee productivity, accelerated sales cycles, and faster global market entry. The momentum for change is palpable, with 71% of leaders now treating workflow transformation as a 2026 priority. Global expansion serves as a major catalyst, named by 33% of respondents as a key investment driver. Meanwhile, 69% of executives expect AI agents to reshape their operations this year—a future that has already arrived for 25% of them.
DeepL CEO Jarek Kutylowski captured the core issue perfectly: “AI is everywhere, but efficiency is not.” He’s right. The failure isn’t in the technology but in its implementation. Most organizations have simply bolted AI onto existing workflows, a strategy doomed to fail. Processes built around people and manual handoffs cannot unlock the potential of AI, no matter how sophisticated the models become. The bottleneck is organizational design, not the tech itself.
Ultimately, the defining factor for success won’t be the size of a company’s AI budget. It will be the courage to rebuild core business processes from an automation-first perspective. Any company serious about achieving a return on its AI investment must audit its translation infrastructure immediately and commit to deep integration of next-generation technologies, not just layering new tools over old problems.
[References & Sources]
- prnewswire.com
- deepl.com
참고문헌
