AI and machine translation can save both time and money – but also entail risks

AI and machine translation – opportunities, risks and best practice

Artificial intelligence (AI) is present in most industries today – and the field of translation is no exception. In a short space of time, machine translation (MT) has gone from delivering substandard results to becoming a serious part of many companies' translation strategies.

Used correctly, AI can save both time and money. But there are also risks: financial, linguistic and confidentiality-related. Therefore, it is important to know when machine translation works and when human expertise is crucial.

What is machine translation?

Machine Translation (MT) is an application of artificial intelligence (AI) that translates text from one language to another using large language models and vast amounts of training data. When you type a sentence into an online translation tool and get the finished text back – that is MT at work.

The result can often look neat and correct at first glance, but it is important to understand that AI does not ”understand” the text. It does not know the context, it cannot determine what tone or feeling is appropriate, and it cannot judge whether the translation is misleading. AI can only mimic understanding based on patterns in previous data.

Raw MT – that is to say, direct machine translation without editing – can work in private contexts or if you want to quickly understand the content of a text. But in professional contexts, it is risky. Furthermore, free online tools mean that you are sharing your data, which in the worst case can be saved or spread further. Free of charge therefore does not mean cost-free.

MTPE – machine translation post-editing

For machine translation to be used professionally, raw machine translation is not enough. The text must always be reviewed and improved by a human. This is called MTPE (Machine Translation Post-Editing).

This is how it works:

Step 1. AI does the heavy lifting
The text is translated quickly and cheaply by an MT engine.

Step 2. A translator takes over
A professional translator goes through every sentence. This usually takes place in a CAT tool (Computer Aided Translation). Here, errors are corrected, nuances are fine-tuned, and terminology and tonality are adapted to the customer's wishes.

Step 3. Quality control
Once the text has been edited, a systematic check is carried out to ensure spelling, grammar, numbers and units are correct, and that the text is consistent and follows established term lists.
Used correctly, MTPE can save both time and money. For simpler texts, such as data sheets or straightforward user manuals between two major languages, the cost can be reduced by approximately 20 % compared with traditional translation. However, for texts where style, tone or accuracy are crucial, post-editing can be more time-consuming – and sometimes even more expensive – than translating from scratch.

MTPE compared to traditional translation

In traditional translation, experienced translators with the target language as their mother tongue handle the entire text from start to finish. This means that the result is not only linguistically correct, but also naturally adapted to the target audience and context. This is the way Språkbolaget has always worked, and we have good experience with this method. The result is very good and maintains a high quality.

With MTPE, AI generates the first draft, and the translator – in this case too, a native speaker of the target language – then steps in to edit and correct. This can be effective for simpler types of text such as basic user manuals, data sheets or internal documents – especially between major languages where there is a large amount of training data for the AI to draw upon.

It is important to know that MTPE does not work equally well for all language combinations. If the amount of training data is small or if the subject matter is very complex, the AI translation often becomes so flawed that the translator in practice has to rewrite almost the entire text. When this happens, both the time and cost benefits of MTPE are lost.

Språkbolaget also has extensive experience of this type of translation, and we always provide a recommendation as to whether we think the text or language combination in question is suitable for using MTPE.

When it comes to texts with high demands on correctness, style and trust – such as annual reports, sustainability reports, legal documents, marketing material or certain types of user manuals – traditional translation is the only option. Here, every word, nuance and phrasing must be completely right.

Benefits and risks of MTPE

Machine translation post-editing (MTPE) can provide several benefits when used correctly:

✔️ Faster process – the translation goes faster for certain types of text.
✔️ Lower costs – especially when dealing with large volumes of text.

However, there are also important risks, both with MTPE and with raw machine translation:

✔️ Lack of context – AI often translates sentence by sentence, which can mean the text lacks a logical thread.
✔️ Inconsistent terminology – the same concept can be translated in different ways in different parts of the text.
✔️ Hallucinations – AI can insert completely irrelevant words or skip important information.
✔️ Apparent perfection – the text may look correct on the surface but contain hidden errors, which makes post-processing time-consuming.
✔️ Confidentiality risks – machine translation almost always takes place in the cloud, which means that data can be disseminated. For sensitive material, this can pose a major risk.

The role of the translator and project manager

Today, a modern translator needs much more than a feel for the language. In addition to expertise in the target language, subject matter knowledge, experience of working in CAT tools, and the ability to manage databases, term lists, and quality controls are required. 

A professional translator must also determine when machine translation can be an aid or when it becomes a risk in order to deliver a text that maintains the right standard.

The role of Språkbolaget is to ensure that all these elements work together. We guide you as a customer by analysing your needs, recommending the right method and putting together a team of translators and reviewers that matches your specific project. Our project managers are responsible for the workflow, quality assurance and deadlines, and handle any unforeseen events. This way, you can feel confident that the translation will be accurate, consistent and delivered on time – regardless of the text type.

When does which method suit?

The factors that come into play are quality, time and cost.

Classical translation
Needed when the text must be completely accurate and build trust, as well as when the requirement for confidentiality is high. Used for legal, medical, financial, annual and sustainability reports, marketing texts, and user manuals with high safety requirements, but also for other texts where a delicate touch is crucial.

MTPE – machine translation post-editing
Combining AI and human review. Suitable for simpler texts where time and cost are important, such as certain instruction manuals, data sheets, price lists and internal documents.

Raw machine translation (MT)
Provides a quick and rough translation without editing. Can be used privately or for an overview when quality and confidentiality do not matter.

AI or human? We help you choose the right method

Machine translation and AI are powerful tools, but they are not a silver bullet. With the right strategy, they can contribute to faster and cheaper projects, but they can never replace human linguistic expertise.

Are you unsure which method is suitable for your next project? Contact us At Språkbolaget we help you find the right level, the right workflow and the right balance between AI and human expertise.