How long will humans be needed for translation?
AI dominates the translation business, but for high-stakes work, humans are still needed.
French rain idioms pose a challenge for AI translators, according to François Yvon, a research director at the French National Centre for Scientific Research. While straightforward weather descriptions are low-risk, more complex topics present higher stakes for machine translation. Google Translate handles one trillion words monthly, with DeepL, Microsoft, Claude, ChatGPT, and Gemini among other competitors.
Machine translation costs a quarter of traditional methods for encyclopedia articles, which are crucial for trust and citation. As AI translation becomes more prevalent, the question arises whether it can replace human translators entirely. While AI output is often accurate, errors can arise, particularly in specialized fields like legal and medical translation.
For instance, AI mistranslated a plaintiff as a requester and a wife as a verb, causing significant issues. Moreover, errors become harder to detect as AI translations sound more natural. In legal and medical contexts, inaccurate translations can have severe consequences, such as delaying detainee releases or causing fatal dosage mistakes.
While AI translation is now the right choice for many use cases, the question remains whether it can be relied upon to the point of removing human translators completely.
Written by urgent.news from BBC Technology's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
