AI and Communication – Can Robots Change Language?

Have you ever stopped to think, when chatting online, whether you are communicating with a real human being, or a robot? Increasingly, companies are investing in artificial intelligence to handle simple communication, and strangely we are normally none the wiser! What is it that differentiates the communication style of a human being and a robot? Why should companies continue to invest in humans to write their content, provide translation services, and handle customer service concerns?


Additionally, has the use of artificial intelligence in communication altered the way we read, or understand language? In previous posts we have looked at the way the English language has changed over time, and the way the European Union’s English is somewhat of an English of its own, due the nature of so many English as a Second Language speakers speaking together over a few decades! Surely, robots learning language and pumping it back into the internet has had some effect on the way we write and perceive online communication.

Artificial Intelligence (AI) is defined as the ability of a machine to perform tasks associated with intelligent beings. In this case, “speak” or write intelligently when prompted. For example, think of a bot who can answer simple customer service questions, or a voice on the telephone who can direct you to the appropriate operator. The idea is that the more the algorithm interacts with real humans, it can actually teach itself new tricks, and become increasingly more intelligent. So, theoretically, the more humans are interacting with bots on the internet, the more intelligent said bots are getting!

Programs have been developed that actually engage in natural language processing (NLP), but humans are still incredibly important. While many companies are excited about the cost-savings associated with NLPs, more companies still are realizing that we still need humans to determine context. A computer knows the definition of a word, but most are still unable to understand the meaning of words within a larger context. This is where the Russian bots that were all over Facebook and Twitter ran into problems – certain distinctions just have to be made in order for language to seem real. That, and typos, as we’ve mentioned before!!

While some companies, like Marbella-based Ravenpack, are banking on humans sitting in cubes to teach their computers what different words mean in certain contexts to eventually render they, the workers, obsolete, we think there is something to keeping a little nuance and subtlety in our language, even if we are still at the computer all day. While science fiction may allow for robots to be poetic, there is still a long way to go before computers can truly understand the complexities of language. Then when it comes to translation and keeping things in context, humans are, amazingly, the only beings on the planet so far who can get it right. And even we fail miserably, sometimes (see our article on translations in China!).

So yes, robots are adjusting language, perhaps improving its efficiency (think of how helpful Grammarly is to check your typos for you), but sometimes there are small errors that only a human eye can catch. While the robots are always improving, we remain one step ahead. That doesn’t mean we shouldn’t use AI to our advantage and to boost efficiency, but it does mean that the translation and interpretation industry will continue to remain a strong field where real humans are incredibly important. Be sure that any agency you work with is not simply outsourcing your documents to firms that rely solely on AI and bots, as you will definitely pay the price when it comes to revisions!