While the tools often take an individualized approach to language training, instructors say they can be helpful in a classroom setting if done right, with tailored audio examples, class activities and student feedback filling some gaps in traditional methods.
“I would say that it's something all language instructors need to have a decent grasp on,” Nicholas Swinehart, managing director of instructional technology at the University of Chicago Language Center (CLC), said. “Not necessarily for their own use, but to understand the ways that learners might be using it.”
Not all professors have taken Swinehart’s suggestion. A 2024 study from the University of Iowa asked 100 language educators about their AI use, finding that only about 20 percent had applied AI to the classroom.
The core components of language education have not changed, according to Karolina Kopczynski, a professor at American Public University who has taught languages since 1999. She said effective language instruction has always meant helping students master listening, reading, writing and speaking. Earlier teaching methods were more rooted in translation and textbook work, and as technology has evolved, so have instructors’ approaches. Most classrooms will employ multimedia materials like videos of TV shows and recorded conversations, Swinehart said, rather than working solely from a book.
Popular consumer-facing platforms like Duolingo and Babbel take a gamified approach, emphasizing short lessons, repetition and steady rewards. These tools track user progress over time, but still largely guide users through lessons in a pre-defined manner.
With AI, tools are better able to support speech recognition to promote conversational practice. Apps like ELSA Speak and Speechling focus on pronunciation and speaking, using voice input and feedback to simulate conversation. Platforms like Lingvist market themselves as explicitly AI-powered, offering adaptive vocabulary decks and industry-specific language learning tracks.
Even AI tools that are not developed specifically to assist with language instruction can be helpful for language teachers.
CLASSROOM CASES
Swinehart said one of the more common uses of AI is curriculum support. For example, an instructor can use ChatGPT to generate 15 classroom icebreakers.
“For any language teacher, something like that helps bring fresh ideas,” he said.
More focused applications involve generating content in the target language itself. Instructors can prompt large language models to produce short texts or dialogues that emphasize specific grammatical features, vocabulary or topics of interest — materials that previously had to be written by hand or sourced from textbooks and media archives.
For example, Ilka Kostka, academic director of Northeastern University’s Global Pathways program who teaches English language courses to multilingual international students, said in a 2024 webinar that she uses AI tools to create customized dialogues focused on small talk. Kostka also uses AI to simplify difficult text materials so they're more accessible to beginners. Swinehart said this is especially valuable for instructors of less commonly taught languages, for whom curricular and multimedia materials are more scarce.
On the student side, AI-based tools provide variety. Kopczynski said students enter language courses with different goals and varying degrees of prior exposure. Some may want conversational skills for travel, others for professional settings. AI-powered tools are often buffet-style, allowing students to focus on areas most relevant to them.
AI tools with voice capabilities can also help students practice speaking outside of class sessions.
“There's a positive potential in terms of having it as a conversation partner,” Swinehart said.
Research suggests incorporating these methods can increase learner motivation and engagement. A meta-analysis of research on AI in language learning published this year in the International Journal of Technology in Education found that tools can increase a learner’s motivation to practice language skills. The study found that learner engagement persisted beyond an initial spike attributed to the novelty of AI tools, potentially due to personalized experiences, real-time feedback and a judgement-free practice zone.
EVALUATING AI TOOLS
Fifteen or 20 years ago, Swinehart said, finding multimedia materials was a primary challenge for people in her field. Now, instructors are more likely to feel overwhelmed by the number of AI-driven tools available to them.
At the CLC, professional development emphasizes starting with institutionally supported tools that integrate with learning management systems and offer data protections. Rather than just adopting popular platforms, Swinehart encourages instructors to focus on what a tool enables pedagogically.
Kopczynski takes a hands-on approach to evaluating tools. She has studied nine languages, but always tests platforms first in languages she is not familiar with to mimic the student experience. Then she evaluates the tool’s proficiency in a language she is familiar with, to make sure it remains valuable as students level up.
WHAT AI CAN’T REPLACE
Despite their benefit, instructors emphasized that AI tools must be supplementary.
“There are so many students who say that I started my learning, for example, through Duolingo,” Kopczynski said. “But it can only take you so far.”
AI tools can lack authenticity and sound robotic, and resources remain unevenly developed across languages. Students will ultimately be tested by conversing with other speakers, and real-world practice is essential to developing those skills.
Still, Swinehart said, AI could mean the difference between an adequate language program and a great one.
“If someone had the language pedagogy and knew how to use generative AI in productive ways, in what I would call good ways, I think that the person using generative AI could get a lot farther in developing their curriculum than someone who wasn’t,” she said.
