Will AI replace translators?
It is hard to miss how new technologies have rapidly invaded our everyday life. Lifeless devices have gained human-like qualities - we speak to them, give them commands and expect predictable responses. And indeed - we are used to the fact that Amazon Alexa or Google Nest almost always understand us correctly. What are the implications of the development of Artificial Intelligence (AI) for the profesisonal translation industry? Does the industry have anything to fear and, ultimately, will translation professionals be replaced by machines?
Translation and new technologies
Breakthroughs in new technologies have always been rapidly commercialised and quickly put to use to solve practical, everyday problems - translations are not an exception.
In the 1980s, as personal computers began to become more commonplace, a company was founded in Germany to offer computer software that continues to be very popular among many translators who want to better organise their work, and avoid translating the same phrases and sentences.
Later, with the arrival of world wide web, a number of solutions have emerged which, inspired by the success of standalone computer programs, have offered users similar solutions based web services, without the need to install software on personal computers anymore. This has contributed to even greater flexibility and ease of the software for the translators.
In recent years, breakthroughs have been made in the development of artificial intelligence. Thanks to the introduction of ever faster processors on the market, it has become possible to develop new algorithms, thanks to which machines have begun to cope with text and speech recognition better than at any time in history.
Are translators drivers?
Autonomous trucks are about to appear on our roads, navigated and driven solely by AI. Job cuts for drivers will seem a natural consequence. Does the increasing adoption of AI mean fewer jobs for translators too?
Machines, which make decisions based on the data on which they are trained, make mistakes. Similarly, humans often make mistakes, and the statistics on car accidents only confirm this. The aim of teams developing solutions based on artificial intelligence is not to create infallible machine. The measure of success is to achieve better statistical results for a given problem than their equivalent in the case of humans.
There were 39,890,500 registered motor vehicles in Q4 of 2019 in the UK and there were 27,222 serious accidents. As there are many more registered vehicles than active drivers on the roads, and accidents often involve more than one vehicle, it can be assumed that the percentage of Britons making fatal decisions on the road is at least 0.2%.
Tesla publishes reports on an ongoing basis on the number of accidents caused by faulty autopilot decisions operating in their cars. According to their statistics, in the fourth quarter of 2019, 1 accident was recorded for an average of 3.07 million miles that their cars had driven. It is difficult to compare this statistic to the one presented earlier. However, it does show to some extent level of sophistication and progress that Tesla engineers manage to achieve.
Translator or AI?
Translations by AI have already become a permanent fixture in certain areas of the economy. Multinational corporations use AI to translate customer enquiries and complaints into the language spoken by employees at the nearest customer support centre. These complaints and queries most often relate to fullfilment delays or technical faults. When translating such communications small nuances of language, which a skilled translator would have translated better than a computer algorithm, are of little consequence. Harnessing AI for these types of tasks allows companies to make huge financial savings and customers to receive better and faster service.
The use of AI in the translation industry is on the rise. In many cases however, the task of translating documents into another language is still a human responsibility, mainly when the translated text text has legal implications.
AI and law
At the moment, it is difficult to find examples of countries in the world where translations by AI would be a legally accepted alternative to translations made by a qualified translators. In Poland for example, a sworn translator must appear on the list of translators of maintained the Minister of Justice. Similarly, in Germany, a sworn (or 'authorised') translator must be sworn in by a district court or national court. In the UK, the situation is slightly different, as the translation profession is not regulated. The main requirement is that the translator must certify the translation by his or her own name or company name in the case of business entities.
AI and translator
Although artificial intelligence will not replace the translation professionals anytime soon, it is already having a huge impact on their work. Artificial intelligence enables translators to quickly catalogue documents by type or language. The huge advances made in machine recognition of printed text make it possible to count characters and words quickly, which makes it much easier to price work.
By making maximum use of the latest technological developments in the field of computer science, the process of making certified translations no longer has to take as long as it used to. The certified trnslation of short documents no longer has to take days or hours, but only minutes.
Approved Translations services
Providing a fast and fully professional translation service is the main objective of Approved Translations. Our mission includes the automation of the translation process, so that our clients can experience the fastest and most convenient translation service, with particular emphasis on quality and security. We believe that the best way to improve the translation process is to make the best possible use of new technologies, while making sure that the translation professional always remains at the heart of the process.
We place particular emphasis on the automation of the translation process. After uploading document files using the form for requesting a quote, you will then receive an email with the quote and a link to a page where you can make the payment after registering with us. After making the payment, you will be able to follow the progress we make working on your documents in real time thanks to status updates. We will notify you by email as soon as the work is completed. The finished translation will be available for download and will be sent by post to the address you specify.