When do AI translation tools make sense?

When do AI translation tools make sense and when is it better to let humans handle the job? Artificial Intelligence (AI) has been part of our everyday lives for longer than we may even realise. Most people consciously started using AI in 2011 when voice assistants like Siri came into the picture. From there, AI took off. And today, we could hardly imagine our everyday lives without it: spam filters, search engines, music and video streaming recommendations, driving assistants and more. Even online machine translation services such as Google Translate or DeepL use AI to produce translations for a wide variety of languages – a discipline that had previously been reserved for human translators. But every AI application has its weaknesses, including in translation. In this article, we explain when translating with AI makes sense and when it does not.

Two of the most well-known AI translation tools are Google Translate and DeepL. Google Translate has been around since 2006 and DeepL since 2011. In mid-July 2024, DeepL’s translation AI was optimised to try and make translations sound even more ‘human’. However, the fact that AI translation tools still need to be improved to make translations sound more human shows that an AI-generated translation is not yet up to par with a human-generated one.

But what about generative AI tools like ChatGPT, Gemini and Perplexity? Can they provide better translations? No. ChatGPT and the like are useful in assisting in creative processes, such as brainstorming and research, but they are not to be mistaken for specialised translation tools.

How are people using AI translations today? For example, Google might ask if you want to display a foreign-language website in your local language. Or a social media platform displays a Translate button under captions, often resulting in strange results. Publishing houses in the USA, such as the New York Times, have also been using AI to translate sports news, specifically for recurring sports terms and phrases. heise online, a German news portal, also announced that it plans to have news translated by AI in the future via its content management system InterRed. But both news outlets have their machine translations proofread by humans before publishing. So when does it make sense to have content translated by AI in everyday scenarios?

This means that, for instance, if you get a letter, an email or any other correspondence in a foreign language, you can use AI to grasp its meaning. However, these texts should follow a simple structure. And even then, errors can creep up, which brings us to the next point.

But why is that? Every AI translation tool still makes mistakes, sometimes resulting in a completely different meaning. Before signing a contract, for instance, you want to be sure you know what it says. Professional human translators can guarantee the accuracy of a translation so you no longer have to worry about it.

Official records, documents, certificates, references – any translation of official documents must be certified to submit to a legal authority. Only sworn translators are authorised to certify translations. That means they attest to the accuracy of the translation with a statement and a stamp. Not all translators can issue certified translations. In Germany, only translators who have taken an oath before a German court can legally do so.

Businesses should not be using AI translation tools as they cannot guarantee the accuracy of their translations. In a business context, running the risk of publishing error-ridden texts shows a lack of professionalism. This applies to legal documents, such as contracts and General Terms and Conditions, but also to marketing and website translations. There is also the important aspect of localisation to consider here. Companies usually need translations as part of their localisation strategy. This is exactly where cultural aspects come into play, especially for advertising materials, slogans, etc. Native-speaking translators are familiar with these cultural insights. This allows them to create linguistically and culturally accurate translations.

In some rare exceptions, businesses can benefit from AI translation tools, for example, for product descriptions or anything with basic, recurring text elements. But even then, a professional translator or reviewer should perform a final quality assurance check.

AI translation tools make sense for personal use, but with limitations, especially when it comes to things like contracts. AI cannot provide certified translations of documents, certificates or references that need to be submitted to official authorities.  

Businesses should avoid AI translation unless followed by a human review process.

Tip: If your business produces content or images using AI, you should label them as such. This will help you build trust with your readers, customers and business partners.