Who Translates Better: Human Intelligence or Artificial?
Two experienced translators and three AI systems—ChatGPT, DeepL, and Copilot—went head-to-head in a challenge organized by the Goethe Institute in Athens.
The stakes? Delivering the best literary translations. A professional audience, including professors and experts in the field, was tasked with judging who excelled in this duel between human ingenuity and advanced technology.
Participants were given 10 minutes to translate excerpts from well known authors such as Friedrich Schiller, Max Frisch, and Felicitas Hoppe. The AI systems impressed with their remarkable speed, delivering translations almost instantaneously. However, when tested on the subtle humor in Hoppe’s texts, the technology faltered. The interpretations missed the mark, failing to capture the essence of the original message.
In contrast, human translators Roussou and Asteriou demonstrated a profound understanding of cultural and literary nuances, faithfully conveying the tone and emotion of the texts. “The machine is fast, but it lacks depth,” remarked Professor Christos Asteriou, highlighting the stark differences between the two approaches.
The Conclusion?
The organizers concluded that while artificial intelligence is a good tool for quick translations, it cannot match the creativity, sensitivity, and intuition of human translators.
AI can achieve remarkable things, but it has yet to replace the human touch behind a perfect translation. Rather than eliminating the role of translators, AI offers them the opportunity to work smarter, not harder. Human-AI collaboration is the future, and translators who adapt their skills will remain indispensable in this ever-evolving industry.
