Why is AI so scary for the translation industry?
Machines are expected to overtake all human jobs by 2140. Although it may seem like a science-fiction plot, this is a harsh reality that awaits us all. How to make money in an economy run by robots? Well, we’ve been doing it for a while. Since the appearance of the first conveyor belt in England and the onset of the Industrial Era, we have been moving more jobs every year to the services sector. The First, the Second, and the Third Industrial Revolution took place and displaced many workplaces into the services. However, with the onset of the more intelligent Fourth Industrial Revolution and the rise of AI, this is becoming a big challenge.
The work done in the AI field in the past few decades is now coming to fruition. You may speak to a synthetic voice when calling your bank to cancel your card, your transaction may get declined by a form of AI and the airplane you took last summer to Ibiza was controlled by an AI autopilot, as it has been for the past several decades. Artificial voices are taking over YouTube narration and Digital Assistants are present in every home, sometimes more than one per household.
The sheer speed at which AI has been taking over the services sector is staggering to say at least. However, some services, such as those pertaining to language and translation, have been left intact. Sure, Google Translate has been around for a while, but it is the best at translating bits of language, never entire pieces of text. The translation produced by these online solutions is always understandable but awkward, and this is the reason why so many people still prefer human translation. IsAccurate is a company offering human-only translation. It is obvious that AI has a long way to go to actually become better than humans at translating or even generating certain communicative intentions, such as creativity, humor, or even sarcasm.
Deep Learning Algorithm and Translation
However, with the introduction of Deep Learning algorithms and the increase in computational speed, it becomes obvious why AI-based translators are getting better year after year. The translation is, after all, a USD 40 Billion industry and whoever manages to fully automate it first will gain an incredible head start over the competition, and having Billions poured into the pockets of any company is sure to secure success and funding for future developments of the service.
Human vs Machine Translation
But how well can the AI today outperform, if even, actual translators? Machine Translation Against Human Translation is not a new kind of competition or research if you prefer, but the rate at which machine translation nears human translators says that around the year 2024, the machines will take over. Computers are already around 2% better than humans at recognizing faces and objects, and deep learning algorithms can even say when somebody is lying or what their sexual preferences are.
AI Translation Final Thoughts
All in all, there will always be people who are willing to pay the extra dime to have a human do their translation. The very fact that a translation could carry a stamp saying HUMAN TRANSLATED would speak volumes of the quality of such a piece of translation. After all, the Industrial Revolution displaced, rather than shut down, workplaces, and the stamp HANDMADE nowadays usually comes at a significant markup. It remains to be seen what areas of language translation and production the AI will penetrate most slowly. It is expected that the translators would move to those spheres.
Michael Carr is bilingual. He grew up in a bilingual family and has learned the value of speaking multiple languages early on. He volunteers in a local public kitchen and loves reading in his second language when he has time.