SIPD – Panta Rhei Enterprise

Conversation and AI: Language in the wild?

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. This way, these chatbots are far superior to those of early applications of AI and especially NLP in CALL (Schulze, 2025).

German chatbot. Generated by ChatGPT 5 as an illustration

Is a conversation with a chatbot the same as a human conversation? LLM-based chatbots have been compared to “stochastic parrots” (Bender et al., 2021), parrots that produce plausible utterances by chance. The computational linguists Bender et al. (2021) stated that

coherence is in fact in the eye of the beholder. Our human understanding of coherence derives from our ability to recognize interlocutors’ beliefs … and intentions … within context … That is, human language use takes place between individuals who share common ground and are mutually aware of that sharing (and its extent), who have communicative intents which they use language to convey, and who model each others’ mental states as they communicate (p. 616).

For language learners, this means that it is on them to control and steer the conversation with the machine. What they say – their prompt – determines what the machine says much more so than with a human interlocutor, who will reason about the student’s intention, which is ‘underneath’ the utterance. The machine will calculate the probability of subsequent word forms. The problem is compounded by the chatbot’s attempt to keep the conversation engaging with immediate questions, which makes it more difficult for the learner to steer the conversation. Similar challenges are to be considered in the other direction: the learner’s understanding of the generated text. Chatbots generate plausible texts that correspond to the student’s input because of highly sophisticated pattern matching processes; chatbots do not generate meaning. We, however, are used to decoding meaning from text. So, it is the learner imbuing the text with meaning when reading the machine output. Students are using their world knowledge, their linguistic capital, and their contextual awareness for interpreting the machine’s output. This is normally based on the human assumption that the chatbot “understood” the prompt. However, machines do not understand text in the way humans do – Hariri (2024) prefers the term alien intelligence for AI, because of their way of processing being radically different – they conduct a fast, sophisticated mathematical analysis and then generate a matching piece of text based on their pre-training. This interaction is akin to that with a calculator, which also cannot understand mathematics and can calculate accurately at great speed. Calculators are faster than humans and are more consistent without making mistakes, and so are LLM-based chatbots in both understanding and generating. So, chatbots have the advantage of speed, accuracy, consistency, and task focus – all without fatigue or stress. Yet, they are lacking in emotional intelligence, awareness of the situational context, and personal memory, for example, of similar conversations with the same person in the past.

Individual conversational practice with a GenAI chatbot is feasible and practical within the context of second-language development. Students are exposed to rich authentic language in the process (Schulze, 2025). However, a genuine negotiation of meaning does not take place, because both text understanding and meaning generation are only done by the learner. Clarification requests, confirmation checks, comprehension checks, paraphrasing, and repair – all part of a negotiation of meaning – can only come from the learner and, unless prompted specifically, do not come from the machine. Feedback on communicatively successful utterances, as we find it in the negotiation of meaning in human conversations, is as yet a challenge for GenAI. Its strengths are in being able to process learner input that contains one or more errors. Such errors often get corrected seamlessly in GenAI’s response or in the prompted correction.

This is part of a draft of an article I wrote with Phil Hubbard. 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.
We call this PD sustained integrated PD (GenAI-SIPD). Sustained because it is continuous and respectful of the other responsibilities and commitments teachers have; integrated because the PD activities are an integral part of what teachers do anyway; the teacher retains control of the PD process.

The full article is available as open access:
Hubbard, Philip and Mathias Schulze (2025) AI and the future of language teaching – Motivating sustained integrated professional development (SIPD). International Journal of Computer Assisted Language Learning and Teaching 15.1., 1–17. DOI:10.4018/IJCALLT.378304 https://www.igi-global.com/gateway/article/full-text-html/378304

In many languages – the ones that are well represented on the internet such that they are a good basis for deep learning – GenAI chatbots can be “patient” and “focused” conversation partners for language learners. They can also go beyond simple turns in a written or spoken conversation and generate various text types for learners and teachers alike. Reading texts, quizzes, instructional sequences, lesson objectives, essays, letters, emails, and others can all be generated. This can create the illusion that a chatbot can be a language tutor.

References

Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610–623. https://doi.org/10.1145/3442188.3445922

Harari, Y. N. (2024). Nexus: A brief history of information networks from the Stone Age to AI. Random House.

Schulze, M. (2025). ICALL and AI: Seven lessons from seventy years. In Y. Wang, A. Alm, & G. Dizon (Eds.), Insights into AI and language teaching and learning. (pp. 11–31) Castledown Publishers. https://doi.org/10.29140/9781763711600-02

Uchida, S. (2025). Generative AI and CEFR levels: Evaluating the accuracy of text generation with ChatGPT-4o through textual features. Vocabulary Learning and Instruction, 14(1), 2078. https://doi.org/10.29140/vli.v14n1.2078

Professional development and GenAI

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. Moorhouse & Kohnke (2024) provide initial insights from a group of language teacher educators on this topic. But what about those who have already completed their formal education and are in the language teaching workforce, the millions of individuals across the world actively teaching languages at all levels?

This is part of a draft of an article I wrote with Phil Hubbard. He wrote 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.
We call this PD sustained integrated PD (GenAI-SIPD). Sustained because it is continuous and respectful of the other responsibilities and commitments teachers have; integrated because the PD activities are an integral part of what teachers do anyway; the teacher retains control of the PD process.

UNESCO (2024) has recognized the immediate need for AI competency across the board in education and why it should be addressed.

Al can pose significant risks to students, the teaching community, education systems and society at large…In education, Al can reduce teaching and learning processes to calculations and automated tasks in ways that devalue the role and influence of teachers and weaken their relationships with learners. It can narrow education to only that which Al can process, model and deliver. Finally, it can also exacerbate the worldwide shortage of qualified teachers through disproportionate spending on technology at the expense of investment in human capacity development (p. 13).

So, given this litany of dangers, what do we think language teachers need to know and be able to do to achieve a functional level of expertise so that they can safely leverage the affordances of GenAI to improve rather than degrade language learning processes and outcomes? How can language teachers and language programs support them in accomplishing this goal?

In our position paper, we address these questions by focusing on the importance of understanding the fundamentals of AI and its subset GenAI, as recognized in several AI competency and literacy frameworks. For example, the UNESCO (2024) AI Competency Framework for Teachers states that at the lowest of their three levels of AI competency, “Teachers are expected to acquire basic conceptual knowledge on AI, including: the definition of AI, basic knowledge of how AI models are trained, and associated knowledge on data and algorithms” (p. 30). The Educause (2024) Durable AI Literacy Framework, targeted at tertiary institutions, goes further: “Faculty must grasp the core principles of AI, including machine learning, natural language processing, and neural networks. This foundational knowledge is crucial for understanding how AI operates and what its potential applications are in various academic disciplines.” Other frameworks we discuss, such as those from the International Society for Technology in Education (ISTE) and Paradox Learning, echo this need for teacher understanding.

References

Educause. (2024). AI literacy in teaching and learning: A durable framework for higher education. https://www.educause.edu/content/2024/ai-literacy-in-teaching-and-learning/faculty-altl

Moorhouse, B. L., & Kohnke, L. (2024). The effects of generative AI on initial language teacher education: The perceptions of teacher educators. System, 122, 103290

UNESCO. (2024). AI competency framework for teachers. UNESCO Publishing. https://unesdoc.unesco.org/ark:/48223/pf0000391104

What teachers need to know now about GenAI

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.
We call this PD sustained integrated PD (GenAI-SIPD). Sustained because it is continuous and respectful of the other responsibilities and commitments teachers have; integrated because the PD activities are an integral part of what teachers do anyway; the teacher retains control of the PD process.

UNESCO (2024). The UNESCO AI Competency Framework for Teachers begins with a chapter motivating the need for AI skills and knowledge and showing alignment with the more general UNESCO ICT Competency for Teachers Framework. The document covers six key principles, the organizational structure of the framework, and specifications for the three levels of teacher competency. The six key principles are based on humanistic considerations: 1) ensuring inclusive digital futures, 2) a human-centered approach to AI, 3) protecting teachers’ rights and redefining roles, 4) promoting trustworthy and environmentally sustainable AI, 5) ensuring applicability to all teachers, and 6) lifelong professional learning for teachers. The document also includes suggested implementation strategies and highlights the importance of adjusting the framework to fit local contexts.

The first of the three levels of teacher competency is Acquire. UNESCO states “The overall curricular goal in the ‘Acquire’ level is to support all teachers to reach a basic level of AI competency or literacy required by the teaching profession across varied contexts” (p. 28). Regarding the second level, Deepen, “The overall curricular goal in the ‘Deepen’ level is to support teachers to become fully competent teachers or master teachers in using AI” (p. 33). And for the third level, Create, “The curricular goal at the ‘Create’ level is to empower teachers who have sound AI knowledge and competency to become expert teachers and agents of change” (p. 38).

Educause (2024). Educause, an association of over 2,100 member organizations, describe themselves as ”a nonprofit association whose mission is to lead the way, advancing the strategic use of technology and data to further the promise of higher education” (https://www.educause.edu/about). They have produced what they call a “durable framework” for AI literacy in teaching and learning, with separate competencies for students, faculty, and staff. The faculty competencies are divided into technical understanding, evaluative skills, practical application, and ethical considerations.

The technical understanding competency begins with Fundamentals of AI: “Faculty must grasp the core principles of AI, including machine learning, natural language processing, and neural networks. This foundational knowledge is crucial for understanding how AI operates and what its potential applications are in various academic disciplines.” This is a particularly strong statement in that it moves beyond GenAI to the underlying principles of other AI applications. A second point to touch on here is “Integration into teaching” under the practical application category. The target competency here is presented as “Faculty should be encouraged to integrate AI tools into their daily teaching activities. This includes using AI to personalize learning experiences, manage course content, and assess AI-informed assignments.” There are other valuable suggestions in this document, though it should be emphasized that the target is university-level education, driven by goals of learning subject content rather than by achieving interactional proficiency, as is the case for much of language learning.

ISTE (2024). Widely considered a major leader in technology and education, ISTE has provided technology standards for teachers for decades. TESOL drew on ISTE standards in the development of their framework for English language teachers and learners (TESOL, 2008). We focus here on the ISTE white paper on AI (ISTE 2024). It offers a framework for updating teacher education programs, which they call “Educator Preparation Programs,” or EPPs. The goal of the framework is to provide guidance in rethinking and restructuring teacher preparation. Although not primarily concerned with in-service learning, ISTE’s AI EPP framework is also relevant for teacher educators and institutions that work with in-service teacher professional development.

The framework is built around three major elements “to guide EPPs in their journey to integrate AI competencies into their programs” (p. 1). The first element, Vision, challenges EPPs to evaluate current faculty and teacher candidate attitudes toward AI and tools in use, followed by developing both short and long-term objectives and a process for tracking progress and achievement. The second element, Strategy, includes deepening faculty understanding of GenAI and how it works, modeling GenAI skills for teacher candidates and providing a foundation in GenAI ethics. The third element, Support, covers the importance of each EPP developing a governance structure at the institutional level, “that addresses accountability mechanisms, transparency, application, ethics, and impact assessment, among other aspects” (p. 6). The framework document includes additional details and illustrative examples useful to language programs and teacher educators involved with in-service teachers rather than initial preparation. Other relevant resources from ISTE are available at https://iste.org/ai.

Paradox Learning (Lee, 2023; Lee, 2025). Paradox Learning is a private company providing elearning solutions and training with an emphasis on developing AI literacy for educational clients. They have created a number of AI resources for educators and made them available open access. Pratschke (2024) recommends one of these, AI Toolkit for Educators (Lee, 2023), and we briefly review it here. In the first part, Lee addresses the question of why to use AI in education, noting how it can be used to support students with resources, personalize learning, provide real-time monitoring and intervention, and aid faculty in administration, research, and teaching. The next part looks at AI tools for administrative and workflow automation, research, and teaching and learning. This is followed by a discussion of challenges and limitations of AI in six areas (e.g., unequal access), with a case study to illustrate each. The next and most innovative element of Lee (2023) is a seven-part framework for what she calls AI literacy (primarily GenAI but including other types), covering both teachers and learners: fundamentals of AI, data fluency, critical thinking and fact checking, common AI applications, AI ethics, AI pedagogy, and future of work.

An update of the AI literacy competencies appears in Lee (2025). The framework is similar to the 2023 AI Toolkit but replaces “common AI applications” with “diverse use cases” and adds an eighth category of “assessment”, clearly needed in the education sector. Like the UNESCO framework, it distinguishes three distinct levels of AI literacy competencies, which Lee labels explorer, integrator, and pioneer. For example, in the assessment category at the lowest “explorer” level, Lee proposes the following target competencies: “List commonly used AI tools for assessment. Describe AI assessment tools’ primary functions, such as automated scoring and personalized feedback. Describe the benefits and limitations of AI in assessments, including efficiency, scalability, and inclusivity” (p. 6).

Beyond these four sources for determining AI competencies in education, those interested in a more comprehensive overview of GenAI in education can turn to Pratschke (2024), Generative AI and Education: Digital Pedagogies, Teaching Innovation and Learning Design. An overview of AI issues for language teaching can be found in Edmett et al. (2024), Artificial intelligence and English language teaching: Preparing for the future. We recommend that teacher educators in particular become familiar with the content of these excellent resources.

As is clear from the preceding review, there is a great deal of overlap among AI competency frameworks in terms of the areas they recommend addressing, even though each is targeting a different population. All four point to the importance of knowing the fundamentals of what GenAI is and by extension what it is not. This suggests that teachers who simply jump in and use GenAI without that foundation face unnecessary challenges compared to those who have taken the time to build a basic understanding of GenAI and its range of functionalities for language learning.

As a first step for teachers, we recommend starting with an overview of GenAI to survey the range of issues and options. In addition to the information and insights in the first half of this paper and in the sources noted above, there are many others freely available online in text or video format that provide accessible surveys of GenAI for educational purposes. For example, the Wharton School at the University of Pennsylvania has produced a five-part Introduction to AI for Teachers and Students that is particularly useful because teacher and student perspectives are considered together: https://www.youtube.com/watch?v=t9gmyvf7JYo.

References

Educause. (2024). AI literacy in teaching and learning: A durable framework for higher education. https://www.educause.edu/content/2024/ai-literacy-in-teaching-and-learning/faculty-altl

International Society for Technology in Education (ISTE). (2024). Evolving teacher education in an AI world. https://1818747.fs1.hubspotusercontent-na1.net/hubfs/1818747/2024_ISTE_whitepaper_EvolvingTeacher_Ed_in_an_AI_World.pdf.

Lee, S. (2023). AI toolkit for educators. Paradox Learning. https://paradoxlearning.com/wp-content/uploads/2023/09/AI-Toolkit-for-Educators_v3.pdf

Lee, S. (2025). AI literacy framework for educators & learning professionals. Paradox Learning. https://paradoxlearning.com/wp-content/uploads/2025/03/AI-Literacy-Framework_updated_031325.pdf

UNESCO. (2024). AI competency framework for teachers. UNESCO Publishing. https://unesdoc.unesco.org/ark:/48223/pf0000391104