AI and the Future of Language Education: What Do Your Peers Think Will Happen in 2025?

How is AI reshaping the future of language learning? 9 leading experts in language education share their insights on the biggest shifts coming in 2025 on Language Leaders, the podcast hosted by LearnCube CEO, Alex Asher. Scan the transcript below or tune in for the full episode here.

Welcome to the Language Leaders podcast! I'm your host, Alex Asher, CEO of LearnCube. LearnCube offers the only virtual classroom and whiteboard made specifically for online language education.

In this special podcast for 2025, I've invited ten remarkable language leaders to share their insights and thoughts around “The Key Trends Shaping Language Education in 2025."

With AI continuing its accelerating course, this year promises to bring transformational changes to the world of language learning. We’ll focus on answering three key questions:

1. What trends will shape 2025 and beyond?
2. What are the most exciting innovations in language education this year?
3. How will the demand for language learning services evolve over the next one to two years?

Throughout this episode, you’ll hear from thought leaders in language education including Tim Jordan SVP of Berlitz, Heather Lo HoP of Learnlight, Stacy Richter CEO of Live Lingua, Vincent Gest Founder of WEFIT, Shoera Sels AI Lead of BLCC, Daniel Delgado COO of Twenix, Luka Smythe Founder of Papo app, Gerardo Onega Director of Marketing and Innovation of Vaughan Systems and Miriam Plieninger,veteran edtech consultant.  

To kick things off, let’s explore our first key question:

What trends will shape 2025 and beyond? 

To answer this, I spoke with Tim Jordan, SVP of Product at Berlitz, who provided a fascinating perspective on how AI will be embedded into language learning—without necessarily diminishing the value of learning a language itself.

"I expect that in 2025, AI-powered tools and features will continue to become more embedded into language learning products. In fact, I think they will start to become the baseline of what learners expect. It will be table stakes to have these features included in online language products.

I think even in-person learners will start to demand access to these digital AI learning tools and the platforms that their online counterparts have access to."

However, while AI-driven tools are becoming increasingly common in language learning, the real challenge lies in how they can enhance rather than replace the learning experience. This is where personalization comes in. Gerardo Onega, Director of Marketing and Innovation, of Vaughan, noted the particular importance of flexibility but I’m going to turn next to Daniel Delgado, COO & Co-founder of Twenix, who explains how AI is shaping more customized learning journeys:

"The key trend shaping 2025, in my opinion, will be the integration of AI to personalize and optimize.  Learning journeys.  AI can analyze vast amounts of data to recommend content, monitor progress, and also automate administrative tasks.  However, the role of human teachers will remain essential. 

Learners value human interaction the most for building confidence, addressing cultural nuances,  also fostering motivation.  Elements that AI cannot fully replicate."

This point on the essential role of humans certainly resonates with me, but I am a little biased, running a virtual classroom software for human language teachers.

I wanted to explore trends outside of the impact on education because AI also is likely to impact how are language companies going to acquire new customers. Marketing channels such as SEO and paid ads may also be disrupted by AI. Stacy Richter, CEO of Live Lingua and long-time veteran marketer had some fascinating insights.

“ While the tools and platforms evolve and change, the fundamentals really don't change at all.  One of my favourite quotes is by John Wanamaker, who said, 50 per cent of your advertising budget is wasted. The problem is we just don't know which 50%.

Now what I'm most excited about in Trends for 2025 is how we can leverage artificial intelligence or AI tools like ChatGPT in order to craft the perfect message to deliver to the perfect client.  At the perfect time with the perfect offer. This is what is going to enable us to eliminate 50 per cent of our wasted budget entirely. 

..that really suggests now we need to increase the number of touch points we have out to those people.  So then if we think about whether or not S conventional channels like SEO and email are dead, this really suggests that they're not dead and they'll become more prevalent than ever. 

But also suggests that it's going to be ever more important to develop more touch points across multiple different platforms and increase our exposure and our story to the perfect client with those perfect messages we're able to craft through our AI tools."

This increasing personalisation is so exciting but I’ve got to admit, it’s not all positive news. This personalisation is fueled by increasingly personal information. There are broader geopolitical, economic and technological trends and challenges at play - this year more than most. Miriam Plieninger, an expert education consultant explains…

"In our increasingly connected and AI-driven world, I'm also worried about data privacy, increasing environmental damage through device production and energy use, and at the same time about a growing digital divide, where different social groups have very different access to technology.  Having said that… many companies… truly support heterogeneous learner groups with their products and ease the work of teachers while trying to navigate privacy, sustainability, and social aspects.

There is a huge opportunity that AI will help to further individualize learner paths and learner practice and learner feedback while keeping teachers' time investment manageable."

Clearly, AI is impacting every part of language education, from marketing and teaching, but what are the broader trends in AI?

To gain a clearer picture, I spoke with Shoera Sels from BLCC, who has spent the last 6 years focused on AI, in language education but also more broadly. Shoera, believes that the real disruption will occur, not when the technology is ready or it makes sense to the finance department, but when we actually work out how to improve academic outcomes for our students using AI.

"I do believe that a lot will be possible in the future. Like we have the speech-to-speech-to-text, for example, which has been slowly improving over the past few years, and it will keep on doing that…And it's all these different things that used to be separate that are now coming to a very mature stage and they are also starting to work together.” 

This integration is key. While AI advancements are impressive, Shoera suggests that the true transformation comes from people—those who can find the right balance between rapidly evolving AI technologies and the real needs of learners.

"So I don't really believe that it's AI itself that will, um, disrupt the market or change the markets, um, in a big way. It's actually these people who managed to find the balance between the AI market that is changing very fast. New opportunities are being created all the time. Um, it's very difficult to predict which one will be first, and which one will be next.

But you just know that they are going to be coming. And at the same time, making sure that you are developing solutions that are actually getting value for the people, um, using them. Um, and I think it's Connecting someone who is able to bridge these two, they will have, they will change the market and they will have an impact on how people learn today and, and how, um, yeah, this impact will be made."

As we can see, AI is not the endgame. It’s a powerful tool but is worthless in language education unless it really does help language learners learn faster, better or cheaper than they otherwise would be able to. Shoera noted in our interview that as a language learner, and an AI expert, she felt so many new products feel like the developers or business people have ignored the educators on their teams. They have slick products but confuse or disappoint the language learner. This is something I noted too when conducting interviews for my book LangTech. It’s so common for langtech companies to have the business and product leaders dominate the decision-making process which, at its extremes can lead to commercially successful, pedagogically low-value language products or services. When I asked her about what she was impressed by in 2024, her response was muted."

"In general, I was not blown off my feet by any actual solution, which does not. Implied that I mean that they are meaningless. I think for a lot of solutions, we are still in the experimentation phase, and that's also how we need to look at them...We are working on this and we are still improving this… What I also see is that they are all in silos a lot… You have every solution that focuses on one key aspect that you can maybe improve by learning. But I haven't seen a lot of solutions yet that actually try to connect these different, um, solutions. I think. It's also partly because this one silo hasn't been, um, yeah, optimized as ideally as, as, as they would like. Looking ahead to 2025 and 2026, I hope to see these siloed solutions improve, integrate more educational expertise, and connect with each other. That’s where I believe the real impact will begin: when these tools work together alongside teachers. I still firmly believe in the combination of teachers and technology to provide the most effective training."

Our second big question was:

What are the most exciting innovations in language education this year?

Heather Lo, Head of Product at Leanlight shared an idea I also believe in. Our customers at LearnCube are also asking us for more seamless integration between human-led and machine-led learning.

"I believe the future of language education lies in seamlessly weaving together human-led and technology-driven instruction…The key is to examine the learner's journey and identify where technology can enhance or replace inefficient human involvement. So, for instance, AI tools can activate prior knowledge at the start of an activity, provide instant feedback, and guide structured reflection, while offering tailored production practice that builds fluency.  Incorporating AI also enables organizations to meet the increasing demand for training in culturally nuanced communication, preparing learners for global collaboration. Further, the data-driven insights AI provides allow companies to measure progress and ensure language training delivers measurable value. Now, that said, there will always be a place for live human trainers. The human connection is vital for fostering motivation, cultural understanding, and nuanced feedback. As technology improves, I envision a growing role for AI. In leading conversational and curriculum-based training, human trainers focus on higher-order interactions that require empathy and complex judgment. And by blending these technological advancements with human elements, like empathy and motivation, we can create a more cohesive and impactful learning experience."

Shoera Sels is looking forward to this seamless integration of solutions.

“ What am I looking forward to in 2025, 2026? It's mainly seeing these kind of silo solutions being improved on, hopefully also taking some educational backgrounds into solutioning the solutions and then also integrating them. And to be honest, I think that's where the actual impact will begin when all these solutions work together to help you together with a teacher.”

However, to create effective AI-powered learning solutions there remain many persistent challenges for innovators to overcome.  To understand these hurdles, I turned to Luka Smythe, CTO and co-founder of Papo, an AI language tutoring app, who has deep insights into the realities of AI product development.

"I think the thing with AI products that's quite unique, is that whilst it's really easy to build a cool proof of concept,  it's even more difficult to make something that's actually really great and reliable.  I think the problem here is that general-purpose AI models like ChatGPT, for example,  as impressive as they are, they'll never be quite right for this task because, you know, they're built to handle all sorts of tasks but language learning has some unique challenges and unique requirements and I think those require custom solutions.  If you take speech recognition, for example,  where the models we have today are already incredibly accurately considered, you know, superhuman in their ability to transcribe speech. But the moment you try and use those same models with someone who's just starting to learn the language, you know, someone whose pronunciation isn't fully developed yet,  they completely fall apart. And when that happens, the whole,  the whole product experience falls apart. So with that being said, I think these kinds of technical problems are very solvable,  but I think that solving them is going to require companies to go beyond even using off-the-shelf AI models."

Miriam Plieninger, an ex-Babbel edtech consultant, is able to see broader trends outside of language education.

“ I've mostly seen existing companies looking at potential efficiency gains last year, at potential time and cost savers. And there have been so many startups trying to get on the market with something AI in the learning field,.., there will be a more growing need to differentiate through learning impact. So I'm expecting that more companies will use AI, not just for internal efficiency, but more so for increasing effectiveness for learners and teachers in an evidence-based way.

Tim Jordan from Berlitz suggested that AI in language learning will transform not just the way we learn with generic chatbots but also improve the efficiency of language companies:

“For example, I think we'll start to see models emerge that are developed specifically for language learning, that will hopefully resolve some of the issues we see with Large language models today where they find it hard to stay on task for very long or are just unable to interact with language learning students at the level that's appropriate for them…So this is particularly an issue for learners at lower levels or those at the more advanced levels where the model just sometimes reverts to the average level that it's been trained on.  I'm also excited about the ability of AI to drive innovation. And efficiency across all aspects of a language learning business, not just where it's mostly used today, which is tutoring or custom support...

For example, we can start to leverage AI tools for better and more efficient customized content creation to develop a truly personalized learning experience where possibly the learner could create their own learning activities based on the inputs they provide in real-time we have AI-powered assessment tools…I'm excited for these to continue to emerge, which can provide immediate and targeted feedback.  There's also the potential to start to build out more features and tools that will support and enable our teachers and our instructors to become more effective and improve the teaching and learning outcomes…Milestones in this evolution include refining AI's ability to assess learners progress in real-time, integrating tools seamlessly into corporate workflows, and ensuring that these technologies enhance rather than replace the human touch”

One of the most important questions every language leader has in the back of their mind is

How will the demand for language learning services evolve over the next one to two years?

While uncertainty looms, the future of language learning is still filled with opportunities. Tim Jordan from Berlitz shares why he believes demand will remain resilient, especially in key sectors:

"I think the demand for language learning services will remain strong, particularly in the B2B, government and academic and higher education sectors. The need for people to communicate and to do business efficiently and effectively across borders will continue to drive the demand for language learning globally."

To provide a helpful discussion on the future of language learning, it’s important to understand and look at the different segments that make up this space. Stacy Richter from Live Lingua provides insight into how different types of learners are shaping the industry:

"I see the demand for language services, uh, continuing for steady growth and bringing many new learners to the space.  These changes that I anticipate are actually much more of a further divide among where the different stages of learners will actually find their services.  For example, the early-stage learners will continue to migrate towards more automated channels like language apps and gaming while the more progressive purpose-driven learners who are either intermediate or advanced will most likely rely on live lessons and tutors and use the technology tools like artificial intelligence, the apps, and gaming as supplements to their learning."

I think this segmentation is important to keep in mind. However, one trend that many leaders shared related to an increasing desire for flexibility. Gerardo Onega from Vaughan Systems explains.

"Flexibility is going to be huge. More and more people are looking for language learning options that fit into their lives rather than the other way around. That's why we expect to see a big rise in on-demand learning, where students get to choose when and how they learn instead of being tied to fixed schedules. 

At Vaughan, for example, we've seen a huge demand for Flex 1 to 1,  which lets learners book and provide lessons with as little as 15 minutes' notice.  And honestly, this level of flexibility isn't just a nice extra anymore.  It's what people expect. Professionals, in particular,  want high-quality training without rigid schedules."

A further trend we’re seeing is the growing interest from B2C companies in expanding into the B2B space. We see that at LearnCube too with many of our online schools competing for corporate tenders. Heather Lo from Learnlight offers a compelling perspective on this shift, accelerated by AI:

“The evolution of AI is reshaping the corporate language training market by transforming both learner demand and service offerings. To understand this shift, we must first consider the consumer market. You know,  just-in-time translation apps and devices are reducing the need for casual language learning related to travel or one-off tasks.

So, while personal interest in learning new languages will persist, I anticipate a sharp decline in consumer-driven language learning.  As a result, major players in the consumer market will likely pivot toward the B2B space, which continues to value language skills for professional communication and global collaboration.”

What is happening in the B2B language learning space?

“Current economic pressures are shrinking L& D budgets, driving demand for cost-effective solutions.

And AI can help providers lower costs by automating aspects of content delivery, learner support, and feedback, while simultaneously improving satisfaction through hyper-personalized experiences tailored to individual needs and learning histories.  With business becoming increasingly global, there's a growing demand for training in soft skills like cross-cultural communication and interpersonal.

So it's not just about language proficiency anymore. And AI can help simulate culturally nuanced interactions and train employees to navigate these complex multicultural settings.  Looking ahead, successful providers will need to adapt by blending human-led and AI-driven approaches, creating scalable solutions that deliver measurable results.”

But AI is not the only force driving change in 2025. I had the chance to meet a number of language business leaders at a recent conference in France. There we’re seeing changes in government, budgets and economic uncertainty really impact the demand for language services. One of those leaders, Vincent Gest, founder of WEFIT shared his observations of the changing requirements and demand for language services best:

"In France, we see, with requests for proposals, There have been requests for face-to-face training, so it is coming back a bit. But the face-to-face language training provider needs to be able to accommodate training in many cities, so they need to be a bit global too.

Speaking of the globalization approach, we see that for all the digital training, it seems that today, providing only English is not enough, you need to have at least five languages, also to provide an app, the mobile app is a must.  And we have seen that certain players who come from B2C have started to be very active in the B2B market so I believe that it will continue in this way."

One technology that a number of our experts commented on was the change in perceptions towards real-time translation technology. The rise of generative AI seems to have coincided with the maternity of natural language processing, in particular speech-to-text technology. We’ve all seen how fast real-time translation is getting, reducing latency to even better than human speed. The shared fear was, that if you can translate someone’s speech in real time, couldn’t you just rely on translation technology to facilitate communication? It is still very unclear that would be the case as Tim Jordan from Berlitz explained.

"One technology trend I'm cautious about is the rise of so-called real-time translation tools as part of business applications like Zoom or Microsoft Teams.

These will be marketed as allowing people to communicate with each other without having to first learn a common language. But my prediction is that the hype won't quite meet the reality of the actual product experience, which is common in a lot of new technology. Uh, the nuance and context needed to communicate in a professional or business setting just won't come close to the effectiveness and return on investment, uh, that a learner would get by actually learning a new language."

Shoera Sels from BLCC shares this scepticism and draws an interesting conclusion. 

"You have seen with translators, there has been a period where they struggled and everyone just believes that technology is solving the translation problems. And now you are already starting to see that people are like, Hmm, maybe not. Maybe we do need translators for specific tasks,

And I think the same movement will happen in, learning technologies. Like, there will be a spike where people are Putting a lot of trust in the technological solutions, um, and it will be very popular, but I also believe it will go down again. I think the demand will go back to also more,  classical teaching ways where you're actually in the class together."

As Shoera was talking, I couldn’t help myself from trying to learn from her insights and somehow make predictions on how AI was going to impact language education and when. Shoera gave some great advice that I’d like to finish on.

"Actually how I try to look at it... let's just think that anything is possible because it will be possible if it's not now, it will be in two years, it will be in five years, it will be in 10 years. 

If anything is possible, how do I in an ideal situation, teach languages to my pupils? What? What do they need? And if you have a very clear image, these are the components that I want, and this is the technology that I need, you can start building back towards this.”

Conclusion

I really appreciate these different viewpoints. No one knows exactly how this year will unfold with big unknowns coming from geopolitical, economic and technological sources so it takes courage to voice an opinion.

I want to thank our contributors for sharing their opinions. Tim Jordan of Berlitz, Heather Lo of Learnlight, Stacy Richter of Live Lingua, Vincent Gest of WEFIT, Shoera Sels of BLCC, Daniel Delgado of Twenix, Luka Smyth of Papo app, Gerardo Onega of Vaughan Systems and Miriam, veteran edtech consultant.  Our contributors have been so generous and helpful. I hope you’ve learned something interesting and actionable for your business.

At LearnCube, we’re here to help you embrace these changes with confidence, offering a suite of software solutions designed to empower teachers, improve the efficiency of language school operations and support faster language learning. 

If you found today’s episode valuable, make sure to subscribe to the Language Leaders podcast for more insights from industry experts. Recent episodes include leaders such as Gio Giovanni, President of ELL at Pearson, and Gaku Nakamura, CEO of publicly-listed RareJob.

P.S. Please excuse the occasional spelling mistakes or errors. This is the automated transcription.