Teachers Still Matter

An insightful new publication

AI is not going to replace humans, but humans with AI are going to replace humans without AI

-       Kevin Scott

 

My last blog entry (see here) carried an announcement of a forthcoming publication titled Teachers Still Matter: Foreign Language Teaching in the Age of AI and an extensive guest-written article related to it by its author Prisha Kohli. I am delighted to say that the book has since been published (see picture above) and is now available for purchase online from the following two outlets:

1.     Notion Press (publishers)

https://direct.notionpress.com/in/read/teachers-still-matter/

2.     Amazon (India)

https://www.amazon.in/dp/B0H29RPC9W/

It is also expected to be available soon at other outlets such as Flipkart.

Though the book is related to the teaching of foreign languages in the age of AI (the author is a teacher of German language and litterateur), it is highly relevant to all teachers in all disciplines as well.  This has been the rationale behind featuring it in my blog.

In a personal note, this is what Prisha has to say about her book:

After spending months researching and writing about AI, pedagogy, assessment, language learning, and the future of classrooms… I can confirm one thing:

Teachers still need coffee more than AI. ☕

Very excited to share that my book is officially launched!

Teachers Still Matter: Foreign Language Teaching in the Age of AI

This book began as one teacher asking:

“If AI can generate language instantly… what happens to language teachers?”

Somewhere between lesson plans, research papers, classroom stories, and late-night writing sessions, this book happened. And honestly, writing a book taught me almost as much as teaching itself.

If you’re a foreign language teacher, trainer, educator, language learner, or simply curious about AI and foreign language education — this book is for you.

Readers can expect a balanced, practical, and deeply human perspective on artificial intelligence in foreign language teaching. The book explains complex AI concepts in simple language that educators can immediately connect to classroom realities. It offers concrete frameworks to help teachers integrate AI without losing pedagogical control. Teachers will discover ready-to-use classroom strategies, prompts, activities, and assessment ideas from A1 to C2 levels. The book also explores major concerns such as AI-generated writing, academic integrity, over-scaffolding, and the illusion of learning. Readers will gain clarity on where AI can genuinely support teaching and where human interaction must remain central. The chapters connect established language learning theories with modern AI-supported classrooms in a practical way. Educators can expect reflection questions, classroom scenarios, and implementation templates that encourage critical thinking about technology use. Rather than promoting AI blindly or rejecting it completely, the book presents a thoughtful middle-ground approach rooted in pedagogy. Above all, the book reassures teachers that their role is not disappearing—in fact, it has become more important than ever.

Here is a concise summary of its contents:

Chapter 1 – How AI Walked into our Classrooms

AI entered classrooms quietly through everyday teaching tasks like worksheets, grammar support, and lesson preparation. The chapter explores teachers’ mixed feelings of curiosity, uncertainty, and cautious optimism about AI in education. It introduces a Model, IAMPC - Intentional AI-Mediated Pedagogical Control Model, which keeps teachers at the centre of AI-supported pedagogy. Ultimately, the chapter argues that AI can support teaching, but meaningful language learning still depends on human interaction and teacher judgment.

Chapter 2 – Understanding AI Without Technical Expertise

This chapter explains AI in simple, teacher-friendly language without requiring technical knowledge or coding expertise. It clarifies that AI works through pattern recognition and prediction, not real understanding or human thinking. Philosophical ideas from thinkers like Paulo Freire, Chomsky, and John Searle are used to show the limits of AI-generated language. The chapter reassures educators that pedagogy, empathy, and professional judgment matter far more than technological expertise.

Chapter 3 – Language Learning Theory Meets Generative AI

The chapter connects major language learning theories with the realities of AI-supported classrooms. It examines how theories like CLT, Sociocultural Theory, ZPD, and cognitive load theory still remain relevant in the age of AI. AI is presented as a scaffold that can support learning only when guided by clear pedagogical intentions. The chapter emphasizes that language acquisition still depends on interaction, negotiation of meaning, and human communication.

Chapter 4 – Teacher-Led Design with AI Support

This chapter demonstrates how teachers can intentionally design lessons where AI remains a support tool rather than the driver of instruction. It introduces practical frameworks and planning strategies for balancing efficiency with meaningful learning. Teachers learn how to create AI-supported activities while protecting learner independence and classroom interaction. The central message is that strong pedagogy must always lead technology integration.

Chapter 5 – Vocabulary and Grammar Learning in AI-Supported Classrooms

The chapter explores how AI can support vocabulary and grammar practice through personalization, examples, and feedback. It also warns against overdependence on AI-generated corrections and explanations that reduce productive struggle. Teachers are encouraged to use AI selectively to reinforce noticing, retrieval, and communicative use of language. The chapter argues that grammar and vocabulary become meaningful only when learners actively use them in context.

Chapter 6 – The Illusion of Understanding

This chapter examines how AI can create the appearance of comprehension without genuine learning taking place. Instant summaries, translations, and explanations may reduce the cognitive effort necessary for deep understanding. The chapter proposes reading models and classroom strategies that preserve inquiry, interpretation, and critical thinking. It reminds teachers that confusion, ambiguity, and struggle are essential parts of language learning.

Chapter 7 – The Silence of Perfect Writing

The chapter investigates the growing problem of polished AI-generated writing that lacks learner ownership and authentic voice. It discusses how fluent texts can hide weak linguistic control and limited understanding. Teachers are encouraged to focus more on process, drafts, revisions, and oral defence rather than only final products. The chapter ultimately reframes writing as a developmental and reflective process rather than a perfectly generated outcome.

Chapter 8 – Teaching Speaking and Interaction in AI-Supported Classrooms

This chapter focuses on speaking skills, conversation practice, and interaction in the presence of AI tools. While AI can provide rehearsal opportunities and simulated dialogue practice, real communicative competence still develops through live human interaction. The chapter highlights repair sequences, spontaneity, turn-taking, and negotiation of meaning as central to language learning. Teachers are positioned as facilitators of authentic communication rather than mere providers of language input.

Chapter 9 – Exams, Evidence, and Learning in the Age of AI

The chapter rethinks assessment practices in classrooms where AI-generated work has become increasingly common. It introduces the AI Assessment Scale (AIAS) to distinguish between acceptable support and excessive AI dependence. Teachers are encouraged to design assessments that prioritize process, independent performance, and visible thinking. The chapter argues that validity in assessment now depends on proving learner ownership, not just evaluating polished products.

Chapter 10 – Bias, Ethics, and Judgment in AI-Mediated Teaching

This chapter explores ethical concerns surrounding AI, including cultural bias, misinformation, fairness, and overreliance on automation. It emphasizes that AI outputs reflect patterns in data and may reproduce stereotypes or inappropriate language use. Teachers are presented as ethical mediators who must evaluate AI critically before bringing it into the classroom. The chapter reinforces that professional judgment cannot be outsourced to machines.

Chapter 11 – Inside the Language Classroom of 2047

The final chapter imagines the future of foreign language teaching in an AI-rich educational world. It explores how classrooms, assessment, interaction, and teacher roles may evolve over the coming decades. Despite technological advances, the chapter argues that human relationships, empathy, and pedagogical guidance will remain essential. The future classroom may look different technologically, but meaningful learning will still depend on teachers.

Conclusion – Teachers Still Matter

The conclusion brings together the book’s central argument that AI should remain a tool under teacher-regulated pedagogy. It reflects on the enduring importance of human interaction, judgment, and meaning-making in language education. While AI may transform workflows and classroom practices, it cannot replace the relational and developmental nature of teaching. The book closes with a reaffirmation that teachers remain indispensable in the age of AI.

Illustrations

Profusely illustrated, with as many as eighty diagrams, pictures, charts, cartoons, etc., the book is also rich in visual content. Below is a random sample of just a few of these:

Appendix: Contents of back cover

Blogger on the author’s work

“Teachers Still Matter: Foreign Language Pedagogy in the Age of AI” promises to be a definitive work that language educators will heartily welcome — one that brings rare wisdom and calm to a conversation too often dominated by misgivings, fear and hype. Through its groundbreaking IAMPC Model, AI Assessment Scale, and five-phase operational cycle, it places in the hands of teachers beautifully crafted, highly practical, classroom-ready, tools that restore confidence and pedagogical authority in an AI-saturated world where language teaching is no exception. Built on rigorous research spanning 250 surveyed educators and 25 in-depth interviews, yet written with unmistakable clarity and human warmth, the book speaks directly to the widely shared reality of teaching — nowhere more tellingly than in the unforgettable personal vignette about the AI avatar, which captures the anxieties of an entire profession and then, with characteristic flair, shows a clear way through. This promises to be a landmark contribution to applied linguistics and teacher education alike — a future-proof, deeply inspiring work that proves, with both evidence and eloquence, that no machine algorithm can replicate the irreplaceable human act of teaching.

Using Claude AI without letting it cloud my own judgement,

Dr S N Prasad

(Teacher Educator and Science Communicator)

An AI-edited picture of the author (left) and the blogger. 

AI cannot bridge the generational gap between them!

 

 

 

 

 

 

Beyond the AI Hype

 The Enduring Role of Teachers

 

“Teaching is not the transfer of knowledge, but the creation of possibilities for the production of knowledge.”

— Paulo Freire

Preface to another Guest Article

In the previous article, Learning in the Brave New World of AI, I had explored how artificial intelligence is rapidly transforming the landscape of education and reshaping the very meaning of learning itself. We are entering an era where access to information is no longer the primary challenge; instead, the real challenge lies in helping learners think critically, independently, and meaningfully in a world flooded with instant answers.

The conversation on this theme naturally leads to an even deeper question:

If AI can generate information instantly, what becomes of the role of the teacher?

This question sits at the heart of an upcoming book, Teachers Still Matter: Foreign Language Teaching in the Age of AI.  Its author and this blog’s latest guest writer*, Prisha Kohli (see insert below), has provided this preview of it.  As a language educator and teacher trainer, she does not see AI as the end of teaching. Rather, she believes AI is revealing more clearly than ever what truly makes teaching human.

[*I also have a very personal association with Prisha – she is my grand daughter-in-law]

From the back-cover of the forthcoming publication

AI Entered the Classroom Quietly

Interestingly, AI did not enter education through dramatic institutional revolutions. It entered quietly through teacher survival. Educators across the world began using AI tools pragmatically to reduce workload and manage growing demands. Teachers started using AI to:

• generate worksheets,
• simplify texts,
• create grammar exercises,
• produce quizzes,
• draft lesson plans,
• generate discussion prompts,
• and save preparation time.

Most teachers are not approaching AI ideologically. They are approaching it practically. And honestly, that is understandable. Teachers today are exhausted.

Educational systems often demand enormous administrative labor while providing limited structural support. AI offers efficiency in areas where teachers have long been overwhelmed. In many cases, AI genuinely helps educators reclaim time and energy. However, alongside these benefits, new anxieties have also emerged.

Teachers increasingly ask:

• How do we know students truly understand?
• What counts as authentic work anymore?
• How do we preserve independent thinking?
• How do we assess learning in an AI-mediated environment?
• Where is the boundary between support and replacement?

What is important here is that these are not technological questions. They are pedagogical questions. The real challenge is not whether AI exists. The challenge is whether education can remain intentional in how AI is used.

While these larger pedagogical questions were unfolding globally, I also found myself confronting AI much more personally inside my own teaching practice.

My Own Moment of Panic

At one point, I experimented with an AI avatar platform capable of generating instructional videos automatically. Watching a digital version of myself teach was deeply unsettling. For a brief moment, I genuinely wondered:

Am I looking at the future replacement of teachers?

The avatar could imitate my voice. It could explain grammar. It could simulate instructional delivery surprisingly well. And yet something felt absent. The more I reflected, the more clearly I understood what AI could not replicate.

AI could imitate:

• instructional delivery,
• verbal explanation,
• presentation style,
• and linguistic fluency.

But it could not reproduce the deeply human dimensions that define meaningful teaching:

• emotional responsiveness,
• relational trust,
• classroom intuition,
• contextual judgment,
• ethical sensitivity,
• spontaneity,
• encouragement,
• humour,
• empathy,
• and human presence.

Good teaching is not merely the transmission of information.

Teachers constantly make invisible pedagogical decisions:

• when to encourage,
• when to challenge,
• when to simplify,
• when to remain silent,
• when to push learners slightly beyond their comfort zones,
• and when emotional support matters more than academic correction.

These judgments emerge from human relationships, not algorithms. That experience fundamentally changed my perspective. I stopped asking whether AI could imitate teachers. Instead, I started asking whether imitation itself is enough for meaningful education. It was at that moment that I began to recognize a deeper problem emerging beneath the excitement surrounding AI in education: the growing illusion that polished performance automatically reflects authentic learning.

The Illusion of Learning

One of the most dangerous assumptions emerging in AI-driven education is the belief that fluent output automatically equals genuine understanding. Today, students can generate:

• essays,
• presentations,
• summaries,
• translations,
• reflective writing,
• and even classroom discussions

within seconds using generative AI tools.

The result often appears impressive. The language is polished. The grammar is correct. The structure feels coherent and sophisticated. But polished output does not necessarily indicate learning. This distinction is especially important in language education.

As language teachers, we know authentic learning is rarely neat or perfect. Real language acquisition involves:

• hesitation,
• uncertainty,
• self-correction,
• communicative risk-taking,
• negotiation of meaning,
• misunderstanding,
• and gradual cognitive struggle.

Learning a language is not simply about producing correct sentences. It is about developing communicative competence through repeated human interaction and meaningful use.

A student struggling to express an idea independently often demonstrates far more genuine learning than a perfectly polished AI-generated paragraph. This is because language learning is not only a linguistic process. It is also cognitive, emotional, social, and cultural.

The danger of AI in education is not merely cheating. The deeper danger is that students may begin confusing generated performance with internalized understanding.

A learner may submit an excellent essay while being unable to explain:

• why certain vocabulary was used,
• why a grammatical structure was chosen,
• or how meaning shifts across different contexts.

This creates a serious pedagogical problem. Education cannot simply measure outputs anymore. It must increasingly examine processes of thinking itself.

I remember one incident from my own classroom when the writing theme was: How do you spend time with your family?

One student submitted an exceptionally polished German article filled with advanced vocabulary and flawless grammar. Everything looked perfect — until I reached one particular sentence:

“On weekends, I passionately hunt my family in the mountains.”

Naturally, I called the student for clarification, slightly concerned about both the grammar and the family. After a very awkward conversation, we finally discovered what the student had actually intended to say:

“I enjoy hiking in the mountains with my family.”

Somewhere between AI translation and overconfident vocabulary choices, a peaceful family trekking activity had transformed into something that sounded like the plot of a criminal thriller. The essay was linguistically impressive. The communicative meaning, however, was an absolute disaster.

As these concerns about authenticity and learning continue to grow, many teachers have simultaneously begun questioning their own place within this rapidly changing educational landscape.

Teachers Do Not Need to Become Engineers

Another major concern I repeatedly encounter among educators is the growing fear that surviving professionally in the age of artificial intelligence now requires advanced technical expertise. Many teachers assume that integrating AI into education means they must learn coding, programming, machine learning, or highly specialized technological skills.

I strongly disagree.

Teachers do not need to become engineers.

What educators truly need is not technical mastery, but pedagogical clarity. The most important skills in the AI era remain deeply human ones:

• pedagogical judgment,
• ethical awareness,
• critical thinking,
• instructional intentionality,
• contextual sensitivity,
• and the ability to evaluate learning meaningfully.

In foreign language education especially, this distinction matters enormously.

AI operates through pattern prediction. It generates statistically probable language based on enormous datasets. It can imitate communication remarkably well. However, it does not possess lived experience, emotional understanding, social intuition, communicative intention, or cultural consciousness.

Language is never only grammar.

·      It is relationship.

·      It is identity.

·      It is culture.

·      It is power.

·      It is human interaction.

For example, in Spanish, the distinction between tú and usted is not simply grammatical. It reflects social relationships, emotional distance, hierarchy, politeness, and cultural expectations. Tú is generally used with friends, family members, children, or people with whom one shares familiarity and closeness. Usted, by contrast, is used in formal situations, with strangers, elders, authority figures, or in professional contexts.

A student speaking to a close friend may say:

“¿Cómo estás tú?”
(“How are you?”)

But while speaking to a professor, the same learner may ask:

“¿Cómo está usted?”

Grammatically, both sentences communicate the same idea. Socially, however, they create entirely different relationships.

These forms communicate:

·      respect,

·      intimacy,

·      hierarchy,

·      professionalism,

·      emotional distance,

·      and relational positioning.

AI may reproduce these structures correctly. But it does not truly understand their human significance. Teaching learners when, why, and how such forms are appropriate requires cultural awareness, contextual interpretation, and human judgment. That remains profoundly human work.

At the same time, this does not mean teachers can ignore AI completely. What educators increasingly need is not programming knowledge, but prompt literacy. In many ways, prompt writing is becoming a new pedagogical skill. A poorly designed prompt often produces superficial, inaccurate, culturally inappropriate, or cognitively weak materials. A well-designed prompt, however, can generate highly targeted classroom support materials within seconds.

The difference lies not in technical expertise, but in pedagogical thinking.

For example, many teachers initially write prompts like:

❌ Bad Prompt Example 1
“Create a German worksheet for class 8th.”

This produces vague and often unusable output because the learning objective is unclear.

A stronger pedagogically guided version would be:

✅ Good Prompt Example 1
“Create a CEFR A1 German worksheet for Indian adult beginners practicing separable verbs in daily routines. Include:

• 10 gap-fill exercises,
• 5 speaking questions,
• Kannada transliteration support,
• and one communicative pair activity.”

The second prompt reflects instructional intentionality. The teacher clearly defines:

• level,
• learner profile,
• linguistic target,
• classroom purpose,
• and activity type.

Similarly:

❌ Bad Prompt Example 2
“Explain German grammar topic conjunctions.”

This is too broad and pedagogically meaningless.

A more effective version would be:

✅ Good Prompt Example 2
“Explain the difference between weil and denn for A2 learners using simple examples related to school and family life. Include common learner mistakes and a short practice activity.”

Again, the improvement comes not from technical skill, but from pedagogical precision.

Another common example:

❌ Bad Prompt Example 3
“Make conversation questions for B1 level.”

This often generates random, repetitive, or culturally disconnected questions.

A better alternative might be:

✅ Good Prompt Example 3
“Generate B1-level role-play speaking tasks for nurses preparing for work in Germany. Focus on patient communication, empathy, and formal language use in hospital contexts.”

This is where teachers remain irreplaceable.

AI may generate language.

But teachers define:

• what matters,
• what is appropriate,
• what aligns with learner needs,
• what supports development,

The future of education therefore does not belong to teachers who become engineers. It belongs to teachers who remain intellectually curious, pedagogically reflective, and critically aware of how technology should serve learning rather than dominate it.

Yet, even when students and teachers learn to use AI thoughtfully, a far more difficult challenge still remains unresolved: how do we evaluate learning fairly in an AI-mediated world?

The Real Crisis Is Assessment

Perhaps the greatest challenge artificial intelligence creates in education is not content generation itself. The real crisis is assessment.

For generations, educational systems across the world have relied heavily on polished final products as evidence of learning. Essays, assignments, homework, projects, presentations, and take-home tasks have traditionally functioned as visible indicators of student understanding. The assumption behind these systems was relatively straightforward: if a student could produce sophisticated work independently, then meaningful learning had likely occurred.

But generative AI fundamentally disrupts that assumption.

Today, students can produce highly polished essays, accurate summaries, grammatically sophisticated responses, and even reflective writing within seconds using AI tools. As a result, polished output alone no longer reliably demonstrates independent competence. A beautifully written essay may reveal very little about whether the learner actually understands the ideas, can explain them independently, or could reproduce similar thinking without technological assistance.

This creates a profound educational dilemma.

The problem is not merely that students may “cheat.” The deeper issue is that traditional assessment systems were designed for a world in which producing polished text required visible cognitive effort. AI has now separated product from process. Students may successfully complete tasks while bypassing many of the intellectual struggles through which genuine learning traditionally develops.

This forces educators to rethink a much more fundamental question:

What does assessment actually measure?

In my own work on foreign language pedagogy and AI, I increasingly argue that future assessment must move beyond static products and focus far more on visible thinking processes. The central concern can no longer be whether students simply produce correct answers. Instead, educators must design assessments that reveal how learners think, adapt, communicate, and respond in real time.

This includes greater emphasis on:

• oral defense,
• spontaneous interaction,
• explanation,
• reflection,
• paraphrasing,
• adaptation,
• communicative flexibility,
• and real-time performance.

For example, a student may submit a flawless foreign language essay generated partially through AI support. But can that same learner explain the vocabulary choices orally? Can they paraphrase their own sentences spontaneously? Can they adapt their ideas when the communicative context changes? Can they sustain authentic interaction without technological mediation?

These questions reveal something far more important than surface-level correctness.

The key educational question is therefore no longer:
“Can the student produce an answer?”

The more important question becomes:
“Can the student think independently beyond generated responses?”

This distinction matters enormously because education has never been simply about information retrieval. Human learning is not equivalent to accessing answers quickly. Education is ultimately about developing individuals capable of:

• reasoning,
• adapting,
• communicating,
• questioning,
• solving problems,
• and eventually functioning independently in the world.

Yet competence rarely develops without cognitive effort. Struggle, uncertainty, revision, misunderstanding, and gradual improvement are not obstacles to learning; they are often the very mechanisms through which learning occurs. When AI bypasses productive struggle entirely, students may complete academic tasks successfully without actually developing durable internalized understanding.

This is especially dangerous in language learning, where communicative competence depends not only on recognition, but on active control under unpredictable human conditions.

Ironically, as AI complicates assessment and exposes the limitations of traditional educational models, it is also revealing something unexpected: the uniquely human dimensions of teaching have become more visible than ever before. That is why teachers matter more now, not less. No algorithm can fully replace that deeply human developmental process.

AI Makes Human Teaching More Visible

Ironically, the rise of artificial intelligence may be helping society recognize the true value of human teachers more clearly than ever before. For decades, many people misunderstood teaching as the simple transfer of information from one person to another. In such a model, education appeared replaceable: if information could be digitized, stored, and delivered efficiently, then perhaps machines could eventually assume much of the teacher’s role. But the emergence of generative AI has exposed the limitations of that assumption. When machines can instantly generate explanations, summaries, translations, exercises, and even entire essays, we are forced to ask a deeper question: What exactly makes teaching human?

The answer lies in everything education was always meant to be beyond information delivery.

AI can generate language, but it cannot genuinely care whether a student is discouraged after repeated failure. It cannot truly recognize the silent anxiety of a learner afraid to speak in front of classmates. It cannot sense the emotional hesitation of a beginner struggling to pronounce unfamiliar sounds in a foreign language classroom. Human teachers can. True language learning involves identity, culture, confidence, hesitation, humour, tone, politeness, misunderstanding, repair, and emotional risk-taking. It requires learners to participate in human interaction, not simply produce linguistically accurate output.

Students therefore need far more than information.

·      They need guidance when they feel lost. They need encouragement when progress feels slow.

·      They need constructive feedback that understands not only what is wrong, but why the learner made that mistake.

·      They need motivation to continue despite frustration.

·      They need someone who notices improvement even before they notice it themselves.

Most importantly, they need someone who believes in their ability to grow.

AI can simulate supportive language patterns. It can produce phrases that sound encouraging. But simulation is not the same as genuine relational presence. A machine does not truly invest emotionally in a learner’s development. Teachers do.

Human teachers build classroom cultures that shape how students experience learning itself. They create emotional safety where mistakes become part of growth rather than sources of humiliation. They mediate conflict, encourage participation, manage group dynamics, and adapt explanations based on individual personalities and emotional states. They recognize confidence, hesitation, boredom, curiosity, and frustration through subtle human cues that machines fundamentally cannot interpret with genuine understanding.

In many ways, AI is clarifying the role of teachers rather than diminishing it. As machines increasingly handle routine informational tasks, the human dimensions of education become more essential, not less. The teacher’s role shifts away from being merely a provider of information toward becoming a mentor, designer of learning experiences, ethical guide, motivator, and facilitator of human development.

The future of education therefore is not a competition between humans and machines. It is a reminder that education was always human at its core. And in an increasingly automated world, that humanity may become the most valuable educational resource of all.

Recognizing the enduring importance of teachers allows us to move beyond simplistic debates about humans versus machines.

The Future Is Not Human vs. AI

The future of education is not a battle between humans and artificial intelligence. It is not a choice between traditional teaching and technological innovation. I do not believe the future lies in rejecting AI entirely, nor do I believe it lies in surrendering completely to automation. The future depends on intentional pedagogy — pedagogy in which technology remains guided by human judgment, educational purpose, and ethical responsibility.

Artificial intelligence is already transforming classrooms across the world. Language teachers today can generate reading materials in minutes, simplify difficult texts for weaker learners, create differentiated worksheets, produce vocabulary lists, design pronunciation practice, and even simulate conversational activities using AI-powered tools. Used thoughtfully, AI can become an extraordinary educational support system. It can support differentiation, accessibility, scaffolding, brainstorming, rehearsal, material creation, and feedback. For students who struggle with confidence or require additional practice outside the classroom, AI can offer opportunities that were previously difficult to provide at scale.

Yet the presence of AI also forces us to confront an important question: What is the role of the teacher when information is instantly available everywhere?

My answer is simple: teachers matter more now, not less.

Technology can generate content, but it cannot truly understand the learner sitting in front of it. It cannot fully perceive hesitation in a student’s voice, recognize emotional withdrawal, sense confusion hidden behind silence, or decide when a learner needs encouragement rather than correction. Teaching is not merely the transfer of information. It is the creation of conditions in which learning becomes possible.

I experienced this very clearly in one of my German language classes. A student preparing for a B1 speaking examination used AI tools extensively to generate model answers. On paper, her responses looked impressive — grammatically correct, sophisticated, and polished. However, during classroom interaction, she struggled to answer spontaneous follow-up questions. When asked to explain her own ideas differently, she became hesitant and dependent on memorized patterns. The AI had helped her produce language, but it had not helped her internalize it. At that moment, my role as a teacher became essential. Instead of allowing the student to continue relying on generated perfection, I redesigned activities that focused on spontaneous communication, negotiation of meaning, and real interaction with classmates. Slowly, confidence and authentic control began to emerge. The problem was not AI itself. The problem was the absence of pedagogical regulation.

In another classroom, however, AI became genuinely transformative. I worked with a mixed-level group where some learners struggled significantly with reading comprehension. Using AI tools, I was able to adapt the same German text into multiple difficulty levels within minutes. Stronger learners worked with the original authentic version, while weaker learners received scaffolded vocabulary support and simplified sentence structures. This allowed all students to participate in the same classroom discussion without feeling excluded or overwhelmed. In this case, AI did not replace teaching; it strengthened differentiated instruction under teacher guidance.

These examples reveal an important truth: AI is neither inherently liberating nor inherently dangerous. Its educational value depends entirely on how teachers design learning around it.

That is ultimately the central message behind my upcoming book, Teachers Still Matter: Foreign Language Teaching in the Age of AI. The rise of AI does not reduce the importance of teachers. It clarifies why they matter. Because education has never been only about delivering information. Education is about helping human beings think, struggle, communicate, grow, question, and eventually become independent.

And no matter how sophisticated technology becomes, that human work cannot be fully automated.