When you listen to a podcast, do you read the show notes? Every time? In case you normally don't, you can read them here before you listen: What can generative AI do for language teaching and learning—and where does it fall short? Mat and Phil beam us back to Oxford, Ohio, for Part 2 of their conversations from the 2026 CALICO Conference…

Banner image of the CALICO 2026 conference What are colleagues working on and thinking about when it comes to AI and language education? Phil and I went to CALICO 2026 at Miami University in Oxford, Ohio in July. CALICO is the Computer Assisted Language Instruction Consortium. This professional organization was founded in 1983. At its annual conference this year, there were lots of presentations, panels, and workshops on things AI.

GenAI-SIPD? SIPD? PD? Let's start at the beginning. GenAI, of course, stands for generative AI. Why? Literally because these AI tools generate something. The ones I am most interested in generate text. They have a large language model - LLM - under the hood. This LLM was created by using - what computer scientists call - machine learning. The result of this process is essentially a bunch of more or less probable connections that are stored in an artificial neural network.

Doesn't AI stand for Always Imitation? Machines are faking it. Computers do what computers do. They crunch numbers. And more numbers. Each letter is a number – think of the ASCII alphabet. Each token is a whole array of numbers. At each artificial neuron, numbers get crunched. And what we see — and believe — is a synthetic text. An imitation.

Phil and I are exploring two contrasting? or complementary? perspectives on AI in education by Andreas Horn and Emily M. Bender. How can language educators respond thoughtfully to a technology that is already shaping education. The two texts by Bender and Horn are contributions to a UNESCO (2025) booklet: AI and the future of education: disruptions, dilemmas and directions.

Things are going well. Really well. ... These are sentences one does not hear very often these days. Yet, here in this blog they ring true. And for that I am grateful. Grateful to you. To you the readers. To you the listeners. Thank you. And let me explain. Photo by Niklas Jeromin on Pexels.com

Podcasts are virtual, right? Phil Hubbard and I were not really in Hong Kong. We were in Hong Kong virtually. We were there virtually, and Glenn Stockwell was there really. Time difference and all. In our podcast episode AI and professional development according to Glenn, Phil had warned that we were coming for him. 😉 In episode 005, Phil and I were talking about Glenn's book chapter…

Join CASLT’s first themed Virtual PL Day to connect, collaborate, and explore AI in language education together.

Natural language processing: In 1990 after having completed my teacher training, I stayed at the university for my doctoral studies. And soon after the Deutschmark came to East Germany on July 1, 1990, I bought a personal computer (Figure 10) and found some software that were not yet in use at the Pädagogische Hochschule Leipzig, because the computers had neither been donated nor delivered yet.

Technology and language teaching — So far, I have mainly talked about the language lab and the TV, which were a big part of my own language learning. In my teacher training, the technologies I had learnt about and was tested on in the late 1980s were the slide projector, overhead projector, record player, and a few others. It was only when I began to teach at the Manchester Metropolitan University in England that I wanted to use computers in language teaching.

In the early 2000s, during the dramatic but now almost forgotten outbreak of foot and mouth disease in cattle, I wrote about war metaphors and the issue of projecting control over an ‘out of …

I continued my language learning at the teacher training college (Pädagogische Hochschule) in at least one Russian language class a day for six days a week in groups of six students and during the third year abroad in Kaluga, a city 200km south of Moscow, at the Pedagogical Institute there. My other classes in teacher education were in phonetics/phonology, morphology, lexicology, syntax, text linguistics, history of the language, for both German and Russian, always in this – structuralist – order, each for one semester,

Six months. I looked it up. The last text before this early beginning of a text is from late January. Nothing in-between. I wrote. Just not here. Just not in the morning. I wrote at work. Emails and notes. I worked on a lengthy book chapter.  Which I now have to shorten. That’s what the editor said.  And also ten scripts.

My language learning began early. I was nine years old and started grade 3 in a special class with the extended teaching of Russian (Wikipedia Contributors, 2026e). My peers in other classes would only begin to learn Russian, as an obligatory subject in East Germany, in grade 5. Before that I had not encountered another language but what they called my mother tongue.

Figure 1: Part of an HSPG (linguistic) sign After outlining the qualitative research method of narrative autoethnography and introducing the research data I used, it is time to begin the story. An autoethnographic narrative can follow a chronological or episodic structure. Since each of the three parts of my narrative are more episodic, in that they address themes that are important for the intersection of AI and language education and relevant for teacher education, the overarching chronology is presented here:

This is an excerpt from an early draft of a book chapter. The book will be about language teacher education and GenAI. I am posting these in smaller (mostly) self-contained posts. The posts are numbered consecutively. After they will have all come out, I will link them with each other. Narrative autoethnography “Ethnography is the study of social interactions, behaviours, and perceptions that occur within groups, teams, organisations, and communities” (Reeves et al., 2008, p.

We are starting with the Akan[1] word Sankofa, which means ‘to return, go, and get’ and symbolizes the taking from the past what is good and bringing it into the present to make the future better through the benevolent use of knowledge (Wikipedia contributors, 2026). The concept of Sankofa has also been adopted in North America and elsewhere. In our context, understanding the past of language learning with technology, of language teaching methods and approaches, and of research and development in artificial intelligence (AI) helps language teachers to make sense of today’s generative AI (GenAI) technologies and their impact on the future of language education, if we shape teacher education now.

Sometimes I forget things ... forget things I did ... forget I did something. Occasionally, I post videos and webpages that I found. Some of these resources, some of my memories have something to do with GenAI or AI in general, others don't. This one doesn't. It looks like a contradiction in terms: a written interview (which I completely forgot) on an…

In this podcast episode, Phil tells Mat about his adventures at the TESOL 2026 Convention in Salt Lake City, March 24-27. Phil talked with exhibitors and attended as many AI-focused presentations, panels, and posters as he could to get a sense of how AI, especially GenAI, is being presented by and to professional English language teachers. Here I will just mention one presentation, about which Phil spoke highly:

Finally ... Mat and Phil left their cocoon, you might say. In the first episodes of their podcast, they talked about their own work. Now they are venturing out. This is still on professional development in the context of GenAI, but now it is someone else's perspective. Phil and Mat talk to Glenn Stockwell about his book chapter on "

You are welcome. In 2025, Phil Hubbard and I wrote an open-access position paper on – what we call – Sustained Integrated Professional Development (SIPD) for GenAI. In that article, we outline ten knowledge and skill areas, which I also discussed here in the blog and about which we spoke in our podcast, and seven principles of SIPD.

Let's use two comparisons: 1 — a jackhammer; 2 — a modern electric car. So, if a GenAI tool is like a jackhammer, what would you do? It might be advisable to learn to use hammer and chisel first. And before that, just learn to use the hammer. Develop your skills so that you get done what needs doing and so that you don't injure yourself ...

That's what we called it. The Oral History Project. We recruited 126 immigrants to Canada from German-speaking central Europe, conducted hour-long interviews about their life, transcribed and annotated the entire corpus, and condensed their individual stories and the different histories into 14 book chapters. Schulze, Mathias, Grit Liebscher, and Sebastian Siebel-Achenbach (Eds.) (2022) Germans of Waterloo Region

What impact will Generative AI have on language learning? I talked about this for an hour last summer, using a personal angle. The methodology I employed is called 'autoethnographic study'. I began by constructing a narrative of my professional life as it pertains to language learning, teaching, and technology. I tried to pick the parts that informed and still inform my thinking about the rapidly evolving GenAI technologies.

By training, I am teacher of German and Russian. Starting in 1986 as a student teacher, I have taught German. First, to kids with German as their first language. Since 1991, to university students as their second language. When I started, I thought German would always be a university subject. It had not occured to me this could ever change. That worry came a few years later ...

Hey, Friend, Still thinking about AI. Every week, when I post here, when I work on a paper or a workshop for work, when I prepare for a podcast. The post below is a little over two years old, and I thought it might be interesting to repost it. Thinking about AI. This is something I have been doing a lot…

Tech-Konzerne sehen in Künstlicher Intelligenz eine einmalige Chance - und wollen einander nicht das Feld überlassen. Die Folge sind beispiellose Investitionen in Rechenzentren.

Artificial intelligence (AI), in particular the areas of natural language processing and student modeling (Heift & Schulze, 2007; Schulze, 2008), have played a role in CALL – this sub-area is commonly referred to as Intelligent CALL or ICALL – for almost 50 years. For a small group of researchers, pedagogical issues of contingent corrective feedback and learner guidance and help were often at the center of attention.

Now I don't know whether GenAI is powerful because it could find a remote and small village, when a manual search could not. Or has it become so dominant that you can find a village in Iran only by looking far down in the search results.

As generative AI tools are rapidly expanding in schools, often based on their hypothetical “potential” to enhance learning, these products are posing risks to children’s development, mental health, pr...

Motivation is only a small part of a student's agency (Knight & Barbera, 2018). Agency plays an important role in all teaching-and-learning processes, as students are the learning subject, setting their own goals and/or accepting or rejecting goals set by others for them. Of course, students do not just determine the goals but also the paths on which they reach them.

Learning about GenAI? Phil and I talk about the first four (of ten) knowledge areas and skills language teachers should aim to have now: prompting, ethics, chatbots, and translation. Combining practical examples with theoretical insights, we discuss these knowledge areas and skills in more detail than we could give in our position paper: AI and the future of language teaching: Motivating Sustained Integrated Professional Development (SIPD)

What the LLM? Have you been wondering about that? What is an LLM? A large language model. If you have been following this blog recently, you will have seen posts on a position paper on Sustained Integrated Professional Development for GenAI I co-wrote with Phil Hubbard. We are hoping this will be useful for language teachers. When writing the paper, Phil and I met regularly (on Zoom) to keep ourselves accountable.

In July 2025, I was fortunate to be invited to the JALTCALL conference in Tokyo. JALT is the Japan Association for Language Teaching and JALTCALL is its (very large) special interest group in computer-assisted language learning. Michael Hofmeyr, the local organizer and brilliant host of JALTCALL 2025 interviewed me before the conference and we (mostly me 😉😅😇) talked for a while.

Often assessment focuses on the product and not the process; the submitted essay gets graded and not the process of writing it. Students can generate these material results of an assessed activity – a short essay, presentation, poster, or audio or video recording – more quickly and with a higher degree of linguistic accuracy than they can create them without the easily available GenAI tools (Fawzi, 2023).

When I served as the director of the Waterloo Centre for German Studies, I started the Jacob and Wilhelm Grimm Lecture, the annual flagship event of the Center. For the last lecture that I organized - with a lot of help from many colleagues - we were fortunate to be able to host the historian Timothy Snyder in early 2017.

Language teachers know that learners need to obtain and then to maintain a level of motivation in their second-language learning and language use. Motivation has become more important in the context of GenAI, as the example of recent advances in machine translation show. Commonly available machine translation tools are now similar to GenAI-based chatbots, because they are also based on large language models.

... is a series of recorded conversations on topics at the intersection of Artificial Intelligence (AI), in particular Generative AI (GenAI), and Language Education. We have started using the acronym OAILL, pronounced /owaıl/. With my accent this sounds like the German weil (=because), but it should be more the w in the English well. The people in conversation are listed in the table below.

Teachers and students alike have used or can use GenAI-based chatbots in their first language. They produce texts that at least sound plausible in the commonly taught languages and in some of the less commonly taught languages, if large language models are available. For these languages, GenAI tools can also function as a dialog partner, natural-language search interface, sketch tool, and text adapter to augment CALL pedagogy.

From late 2022 to early 2025, several LLM-based chatbots were released, such as ChatGPT, Claude, Gemini, Qwen, Llama, and DeepSeek. All of them can generate conversational responses with remarkable linguistic accuracy and contextual appropriateness and at almost turn-taking speed. Such chatbots can be prompted to adjust the proficiency level of the output to the learner (but see Uchida (2025) for the challenges with adjusting for CEFR levels) and can stick to the desired topic and only use appropriate vocabulary.

Before 1916, the city of Kitchener, Ontario, Canada, was called Berlin. Ten years ago in 2016, colleagues from the Waterloo Centre for German Studies and I organized a panel discussion that marked the 100th anniversary of that name change. This discussion took place and was recorded in the Kitchener Public Library. Carl Zehrs, the former mayor of Kitchener, was our moderator.

The degree to which generative artificial intelligence (GenAI) has rapidly infiltrated education is unparalleled. Language education has been particularly impacted because GenAI tools process and generate the learning objective of that education, i.e., human language. Language teacher education programs have been faced with addressing GenAI since the public release of ChatGPT in November 2022, and we anticipate that many recent and future graduates will have had some formal education that includes it.

Twitch account for The Distributed AI Research Institute (DAIR).

We propose 10 areas for language teachers to develop fundamental knowledge and skills in. While these do not constitute an exhaustive list, they cover a wide range of the main applications of GenAI for language education. We suggest that teachers devote a few hours to exploring each as soon as possible, reflecting critically on how they may be incorporated in current teaching contexts.

We begin by reviewing four frameworks covering teacher competencies for AI and GenAI in education as a whole – UNESCO (2024), Educause (2024), ISTE (2024), and Paradox Learning (2023, 2025). This is part of a draft of an article I wrote with Phil Hubbard. He was the main writer of this part. In this paper, we are proposing a way in which teachers can organize their own professional development (PD) in the context of the rapid expansion of Generative AI…