Artificial intelligence has become one of the most important forces reshaping the translation market. This does not, however, mean a simple end to the profession of translator. A more likely scenario is a profound restructuring of the industry: some simple assignments will be taken over by automated tools, while specializations requiring responsibility, expert knowledge and quality control will gain greater value.
According to an analysis published by Biuro Tłumaczeń 123tlumacz.pl, the global language services market is still growing in 2025, but its structure is changing. Broad estimates put its value at around USD 71–72 billion, while narrower approaches, such as the Slator report, indicate approximately USD 31.7 billion. The difference results from the varying scope of the market: some studies include written translation, interpreting, localization, subtitling, dubbing and language technologies, while others cover only part of these services. (Biuro Tłumaczeń 123tlumacz.pl)
The market is growing, but traditional translation is becoming cheaper
The greatest paradox of the current transformation is that the world is translating more and more content, but traditional translators are not necessarily being paid more. Automation is reducing the cost of the first draft of a translation. A client who, just a few years ago, would have ordered a translation from scratch now increasingly comes with a text generated by an AI system and expects only verification.
In practice, this means that value is shifting from translation itself to quality control. Translators are increasingly becoming editors, post-editors, linguistic auditors and risk specialists. Their task is no longer only to transfer sentences from one language into another, but to assess whether the text is correct in terms of terminology, law, culture and business context.
According to the analysis cited, Europe remains one of the key markets for language services, mainly due to the multilingual nature of the European Union, the large number of international companies and the extensive circulation of legal, official, technical and financial documents. The Polish market, meanwhile, is highly fragmented: it includes around 1,500 professional translation agencies, approximately 11,000 sworn translators and a large group of freelancers. (Biuro Tłumaczeń 123tlumacz.pl)
Which types of translation are most at risk?
The greatest pressure is being felt in segments where the text is repetitive, simple and low-risk. This applies primarily to general translations, product descriptions, simple corporate documents, FAQs, business emails and basic interface localization. In such cases, AI often generates text that is good enough for quick internal use.
There is also a high level of risk for some technical translations, especially when the terminology is standardized and the documentation has a repetitive structure. A similar process can be seen in subtitling and audiovisual translation, where automated systems increasingly prepare the first draft of subtitles, while humans correct errors.
Legal, medical, pharmaceutical, clinical, literary, sworn and conference translation remain much more resilient. In these areas, an error may have legal, financial, health-related or reputational consequences. The higher the cost of a mistake, the greater the value of human involvement in the translation process.
Post-editing is becoming the new standard
One of the most important developments is the growth of machine translation post-editing, or MTPE. This means that the first version of the translation is prepared by a system, and then a translator checks the text, corrects errors and brings it up to the quality required by the client.
The problem is that post-editing is not always easier than translating from scratch. In specialist, medical, legal or creative texts, checking AI errors can be time-consuming because machine-generated text may appear correct while containing subtle errors in meaning. For this reason, the expectation of large discounts simply because an AI tool was used is increasingly being challenged by translators.
The analysis by 123tlumacz.pl notes that typical rates for post-editing are lower than for traditional translation, but the difference does not always reflect the actual amount of work involved. This is one of the main points of tension between clients, translation agencies and freelancers. (Biuro Tłumaczeń 123tlumacz.pl)
How many translators may leave the market?
It is not possible to reliably indicate one specific number of people who will lose their jobs by 2030. The profession of translator includes many specializations, and data on freelancers is incomplete. Some people may not formally lose their jobs, but they may lose a significant share of their assignments or be forced to change their professional profile.
However, three scenarios can be outlined. In a cautious scenario, 20–30% of translators may lose a significant share of assignments or leave the profession. In a moderate scenario, the problem may affect 40–55% of those working mainly in the general translation segment. In a pessimistic scenario, AI pressure could lead to the departure of as many as 60–75% of translators from the most vulnerable market segments.
The most exposed groups are general translators, people handling simple technical translations and freelancers without a clearly defined specialization. The best prospects, on the other hand, are held by medical, pharmaceutical, legal, sworn, literary and conference translators, as well as specialists in rare language pairs.
Who will survive the change?
The best chances will belong to translators who can combine linguistic competence with industry-specific knowledge. Knowing a language alone is no longer enough of a differentiator, because the first draft of a text can now be prepared by an automated tool. What becomes valuable is the ability to recognize whether that draft is safe, terminologically consistent and appropriate for a specific context.
The most future-proof specializations include translation for life sciences, clinical trials, pharmaceuticals, law, finance, energy, technology, public administration and technical documentation related to product safety. The importance of AI translation audits, quality control, terminology management and validation of automatically generated texts will also grow.
New professional roles may include post-editor, AI translation auditor, terminology consultant, Language QA specialist, cultural localizer or expert in implementing language tools in companies. These will still be language-related professions, but they will be based less on mechanical translation and more on knowledge, assessment and responsibility.
Translation agencies must also change their business model
The change does not affect freelancers only. Translation agencies that have mainly acted as intermediaries between clients and translators may lose part of their advantage. Corporate clients will increasingly implement their own AI systems or use Translation as a Service platforms.
The agencies that survive will be those offering added value: specialization, data security, procedural compliance, quality control, auditing, large-scale project management and responsibility for the process. Regulated sectors will be particularly important, as clients in these areas cannot afford uncontrolled use of public AI tools.
TMS systems, terminology databases, private translation models, project management automation and a documented quality control path will also become increasingly important. The translation agency of the future will be more of a technology and compliance company than a traditional point of file transfer between client and translator.
Sworn translations remain more resistant
Sworn translation is a special segment. Its resistance to automation results not only from linguistic difficulty, but above all from legal regulations. A certified translation requires a person with proper authorization, a signature, a seal or appropriate electronic authentication.
AI can support the work of a sworn translator, but it cannot independently assume responsibility for a document. Courts, public offices, universities, the Polish Social Insurance Institution, prosecutors’ offices and public institutions will continue to require translations prepared and certified by an authorized person. This is why this segment may change its working tools, but it will not be easily replaced by automation.
Digitalization may, however, change the form of authentication. Qualified electronic signatures and digital document circulation may play an increasingly important role. This does not mean, however, that the requirement of professional responsibility on the part of the translator will disappear.
