5 Reasons Why You Should Be Careful About Machine Translation

Machine translation (MT) usually refers to using algorithms and machine learning (ML) models to translate natural language text from one language to another without human intervention. The most common MT examples include but are not limited to Google Translate, Bing Microsoft Translator, Amazon Translate and DeepL.

With the rapid development of generative artificial intelligence (AI) and ChatGPT, many industries face unprecedented challenges, and the translation industry hasn’t been spared. Taking efficiency and cost into consideration, more and more business clients tend to use machine translation to complete their projects.

However, is it always a wise choice? Here are 5 reasons why you should be careful about machine translation:

Cultural Accuracy: Every culture possesses unique lexical terms, slang, and colloquialisms that machines haven’t shown the capability to translate yet, inaccurate translations may lead to poor interpretation of your brand, vision, market position and business strategies.

Human Touch: Human translation goes through a time-tested process of multiple editing and proofreading to ensure that the translation isn’t only grammatically correct and readable, but always enhanced for the target audience. On the contrary, machine translation can only generate simple, toneless text, it’s fast and budget-friendly, but it can never be intriguing.

Flexibility: Language is constantly evolving, one single term may have entirely different meanings in different contexts, let alone phrases, sentences or even longer paragraphs. MT tools can only generate translations based on the known corpus, they can’t predict and correct specific grammatical and cultural errors like human do.

Layout: Good translation takes time, so does formatting/layout. When we assess the quality of translation, formatting/layout also palys an important part. Unfortunately, almost all MT tools can’t handle this properly, they just ignore it or put some illustrative texts instead.

Confidentiality: As a responsible language service provider, we should never disclose customer information to any unauthorised third party. But as far as I am aware, some MT tools, especially ChatGPT, may collect and store different kinds of user input, which can be a great security risk for businesses.

In short, it’s OK to use MT tools in less important content (i.e., content that does not require translation precision and extensive copywriting). But when it comes to business documents or audience-facing content, there is nothing can beat human translation.