Gabriel Guillén spoke at WashU about AI and language education (Rachel Benitez-Borrego | Staff Photographer).

Gabriel Guillén, a professor at the Middlebury Institute of International Studies at Monterey, discussed language learning in the age of artificial intelligence as part of the Ginger Marcus Foreign Language Learning Speaker series on April 2.

The talk, Language Education and AI: What We Teach When We Stop Being Afraid, covered what differentiates humans from animals, why humans still feel compelled to learn new languages even in the age of AI, how AI is different than other tools throughout history, and how AI can supplement language learning rather than diminish it. 

At the start of his event, Guillén took an online poll of audience members, most of whom were foreign language professors, on how comfortable they felt with AI. Most of the audience fell between “AI cautious” and “AI curious,” and two professors even felt “AI committed.” 

What makes humans unique, Guillén then told the audience, is generative language. 

“We can create an unlimited set of meaning with a limited set of sounds.” 

However, without passing on language between generations, humans could not have persisted as a species, Guillén said.

“We need to teach the language. If we don’t teach the language to our kids, then it doesn’t matter, right? So one can say that teaching language is what makes us human.”

Even in the age of AI and tools that have predated AI like Google Translate, there is an ever-increasing amount of people who want to learn more languages, Guillén showed through data graphs. Part of this can be attributed to an increase in migration due to climate and economic reasons. 

In the 1970s, there were 250,000 students taking Spanish in American classrooms. Today, that number is half a million, Guillén showed on a graph. 

Guillén told the audience that using AI to communicate with people who speak different languages cannot truly replace real communication in the same language. 

“We are language, and it allows us to create a new identity, create belonging, trust with other people when we move to other places, or when we welcome people from other places.”

To show how AI is different from other tools, Guillén had the audience try to pair tools throughout history with what they “expanded.” The telescope was an expansion of the human eye, the hammer an expansion of hands, and the book as an expansion of our memory. He then asked the audience what they thought AI was an expansion of, or if it was another thing entirely. 

A professor in the audience responded that AI is different because it creates something that we don’t have without acknowledgement of authorship. Even if we are the ones prompting the AI, what it produces is still not from our own thinking and what we know, but rather from a collective knowledge programmed by other people, the professor explained. 

Another professor remarked that AI is unique because it can replace some human roles, serving as a personal assistant. 

So, Guillén asked, how should professors adapt assignments to still encourage learning from their students? 

“Maybe we can focus on local or recent events, audio or video assignments, and focus on process over product,” Guillén proposed. 

Through incorporating AI into curricula and telling students on which steps of assignments they are allowed to use it, professors can teach students not only how to use AI effectively, but also how to spot limitations or mistakes it makes, Guillén said. 

“I’m creating a new AI policy. I’m sure many of you are doing this right now. Mine is based on agency, the productive struggle, transparency, and integrity of the voice,” he said. “AI policy should be a process, not just something like a contract.”

Something that AI cannot replace is immersion, Guillén said. Even with AI chatbots that mimic conversations for language students to practice speaking, it is impossible to replace the study abroad experience. 

Lecturer in Italian and audience member Paolo Scartoni, who responded “AI cautious” to Guillén’s initial poll, told Student Life that he attended the talk to learn how his colleagues are dealing with AI. 

“I try not to use AI in my classes, so I was here to see if somebody could change my mind,” he said. 

After the talk, Scartoni said he still remains skeptical about whether AI does more good than harm to students during the learning process.

“It’s my sense that we’re asking AI to make the process more efficient, in terms of making it quicker. And I’m a big fan of taking things slow. I think there’s value in enjoying the journey. And I don’t think there’s always value in finding shortcuts,” Scartoni said. 

Use of AI in the Romance Languages and Literatures department varies among professors, Scartoni said.

“I think it’s great that different people are using different methodologies and exposing students to different formats of instruction.”

However, Scartoni said he is hesitant to add AI into his teaching because he fears it will remove an essential part of learning: making mistakes.

“The most beautiful things you learn little by little, and it takes a long time. I think our biggest success as language instructors is when we make students passionate about learning that language, because it’s a journey that does not end in four years.”

Scartoni, whose first language is Italian, has noticed a difference in his English language improvement since beginning to use AI, even just for writing emails.

“I’ve noticed that before AI, my English improved faster because I had to pay attention to friends and colleagues who gave feedback to me, whereas now I just pass it through AI, and I don’t even pay attention, because I know that I have it there whenever I need to use it,” he said.

In the classroom especially, instructors have a responsibility to protect students’ thinking and learning and to consider what using AI might do, Scartoni said. 

“Some of my concerns are the fact that behind these AI tools, there’s an algorithm, and behind the algorithm there’s people, and that’s scary because we are delegating our thinking, our learning, our writing, to some faceless people who [are] not in learning institutions. They are in private companies. Their goal is to make money,” he said. “We are asking you to exist in a space that is designed by someone with specific interests that we know nothing about.”