Why AI Is Not the End of Language Learning (and the Many Ways It Can Help You Get Fluent)

There is no denying that artificial intelligence (AI), large language models (LLMs), text-to-speech (TTS), translation tools, and chat bots have changed how we live, learn, and work. And they will certainly continue to disrupt industries, cultures, and curricula for years to come.

The rapid advancement and adoption of grammatically accurate AI tools like ChatGPT and visually accurate deep fake videos like HeyGen have led to a slew of sensationalist articles, bold claims, and depressing predictions.

For example, a recent Atlantic article titled “The End of Foreign-Language Education” has been making the rounds this week (and has been forwarded to me by friends and family concerned with the future prospects of my language-based career). The article begins with the author, Louise Matsakis, questioning her efforts to learn Mandarin Chinese after watching a HeyGen avatar of herself speaking "perfect Chinese":

"I’ve been studying the language on and off for only a few years, and I’m far from fluent. But there I was, pronouncing each character flawlessly in the correct tone, just as a native speaker would. Gone were my grammar mistakes and awkward pauses, replaced by a smooth and slightly alien-sounding voice. ... By merely uploading a selfie taken on my iPhone, I was able to glimpse a level of Mandarin fluency that may elude me for the rest of my life. ... its language technology is good enough to make me question whether learning Mandarin is a wasted effort."

Fortunately, her concerns are not warranted.

Sure, such AI-powered language tools are impressive. And I can certainly empathize with Louise's envy (it takes us humans years to internalize what a computer seemingly does in seconds). But don't underestimate the power of the human brain. With enough exposure and practice, you will reach fluency in Mandarin (or any other human language). And you won't sound like a robot or need to rely on a digital device to communicate.

Here are three reasons why AI is not a harbinger of death for language learning, and my rebuttals to some of the false leaps of logic presented in the Atlantic article.

Classrooms Are Not Necessary to Learn Languages

The article shares some pessimistic statistics about reductions in foreign-language studies at schools in many industrialized counties:

”…the number of students studying foreign languages in the U.S. and other countries is shrinking. Total enrollment in language courses other than English at American colleges decreased 29.3 percent from 2009 to 2021, according to the latest data from the Modern Language Association, better known as the MLA. In Australia, only 8.6 percent of high-school seniors were studying a foreign language in 2021—a historic low. In South Korea and New Zealand, universities are closing their French, German, and Italian departments.”

The author then assumes a correlation between the rise of AI-powered language tools and this reduction in language education:

“… it’s clear that people are turning away from language learning just as automatic translation becomes ubiquitous across the internet.”

This seems alarming on the surface, and it is indeed a shame that there appears to be a downward trend in institutional interest in—and emphasis on—foreign languages.

However, just because rates of foreign-language education are decreasing in schools, it does not automatically follow that rates of foreign-language learning are decreasing in the world.

First of all, classrooms are not the only place to learn a language. In fact, in many ways, they are an awful place to learn. At school, languages tend to be treated as academic subjects that can be systematically taught, memorized, and tested.

The truth is that language acquisition is an innate drive more akin to learning to walk than learning math. We learn to walk (and speak!) by trying, falling, and getting back up again and again. It is a matter of active physical practice and subconscious internalization, not conscious study and memorization.

Second, today’s language learner has unprecedented access to language learning resources, tools, and tutors (both human and AI-powered). Your address, income, and schedule are not the barriers they once were. Got 15 minutes during your lunch break? Fire up the italki app on your phone and practice speaking one-on-one with a native speaker. Or you can even get 24-7 personalized practice with an AI-powered teacher using Teacher AI. This is but one of many cases where AI tools don’t replace humans, but rather supplement them.

So don’t despair if your high school or university doesn’t offer classes in the target language you want to learn (or hope your children will learn).

You don’t need a class. You need exposure and practice. And these two essential ingredients are actually easier to acquire than ever thanks to the very AI tools that are supposedly killing the need for language learning!

AI Tools Can Help You Immerse & Practice

Getting fluent and literate in a language requires an immense amount of input (listening and reading) and output (speaking and writing). A large proportion of this exposure and practice will naturally be provided by other humans, whether you are practicing speaking with an online tutor, getting your writing corrected, watching a film, listening to a podcast, etc. And these are all excellent ways to get the immersion and practice you need.

However, such human-dependent resources needn't make up your entire language-learning toolbox. AI-powered tools have many advantages:

• 24/7 availability. Computers don't sleep, so AI-powered tools like Teacher AI are available whenever you want to practice. Tutoring platforms like italki (albeit extremely useful in their own right), require you to schedule in advance and meet up at a specific time.

• Lower pressure. Trying to communicate with a native speaker can be intimidating, especially in the early stages when your grammar and vocabulary are still limited. AI tools remove potential performance anxiety and reduce the social stakes to zero.

• Personalized practice. Tools like ChatGPT can point out mistakes in your vocabulary and grammar, rephrase lengthy passages in simpler language, provide back and forth conversation practice, answer questions, and fill in gaps in your linguistic or cultural knowledge.

Languages Are about Real-Time Connection, Not Translation

It’s true that some AI-powered translators can now provide rapid, (mostly) accurate translations between different languages. But we are a long way from a Star Trek-style “Universal Translator” that lets you speak your language as your interlocutors hear what you are saying in their language with no delay.

Instead, you have to say a sentence or two into a device as the other person waits patiently. They then hear or read the translation. And then repeats the process the other direction. This may help monolinguals communicate, but does it help humans connect?

I argue no. Having to rely on a device creates a literal physical barrier between parties. And the time delay creates artificial, laborious exchanges.

This is one point on which I agree with the author. Those who give up on language learning in favor of translation tools will lose more than they gain:

“As the technology becomes normalized, we may find that we’ve allowed deep human connections to be replaced by communication that’s technically proficient but ultimately hollow.“

We’ve already seen this with social media, texting, email, etc. in years past. New digital tools often end up overtaking deeper, organic forms of connection.

But! It’s important to remember that we still have a choice!

• Just because texting has become normalized, it doesn’t mean we can’t still call people.

• Just because email is the default form of written communication today, it doesn’t mean you can’t send a handwritten letter.

• And just because Google Translate can render your thoughts in another language, it doesn’t mean you can’t or shouldn’t learn to do so yourself.

Spoken fluency in languages is part of what makes us human. And reaching conversational fluency unlocks rapid, authentic, intimate human-to-human connection that no technology will replace any time soon. It may someday, but I think it’s farther away than what many tech evangelists and doomsayers claim.

And such a level of fluency is not only possible (despite the author’s fears), but it’s all but guaranteed if you do the right things. Again, getting sufficient exposure and practice. NOT sitting in a classroom.

So no, AI will not replace the need for learning foreign languages. Nor the many benefits fluency provides. On the contrary, these tools make it easier and more efficient to acquire languages anywhere, anytime using the immersion-based approach I advocate in my Quickstart Guide, books, and courses.