Brief Reports in Language Research

 

Brief Reports in Language Research, 1, 260101 (2026)

https://doi.org/10.65553/BRLR.260102

 

 

Dancing with Shackles or Going with the Flow? Teacher Agency and Generative AI Integration Among Tertiary Teachers in Hong Kong and Macau

 

Junjie Gavin Wu*

 

Macao Polytechnic University, Macao SAR, China

 

Abstract

Since the release of ChatGPT in November 2022, generative artificial intelligence (GenAI) has rapidly entered higher education, yet how in-service teachers exercise agency in integrating it remains under-researched. Drawing on an ecological view of teacher agency, this qualitative study examines the practices and challenges of tertiary teachers in Hong Kong and Macau, two technologically advanced but institutionally distinct contexts. Twenty in-service teachers (10 per city) from multiple disciplines took part in semi-structured interviews, submitted lesson plans, and answered open-ended questions. This report presents preliminary findings from two information-rich cases. Chan (Hong Kong) worked within strong institutional support and peer pressure, integrating GenAI extensively while feeling “forced to go with the flow.” Lau (Macau) faced an absence of policy, training, and collegial exchange, yet integrated GenAI through self-directed learning, describing herself as learning “to dance with shackles on.” Neither case fits a binary of proactive versus reactive agency; instead, both teachers moved between the two as institutional affordances, past experience, and future aspirations interacted. The study argues for a temporal, relational understanding of agency and highlights implications for institutional GenAI policy and teacher professional development.

 

 

Keywords: generative AI; teacher agency; ecological perspective; higher education; Hong Kong and Macau


 

 

 

Introduction

The public release of ChatGPT in November 2022 marked a turning point for education. Generative artificial intelligence (GenAI) can produce text, feedback, and learning materials on demand, prompting universities worldwide to reconsider teaching, assessment, and academic integrity (Kasneci et al., 2023; Watson Todd, 2026). Institutional responses were uneven: a review of the world’s 50 top-ranking universities found that just under half had published GenAI assessment guidelines (Moorhouse et al., 2023). In language education, GenAI has been framed as both an affordance and a disruption that calls for new teacher competencies and a rethinking of where agency lies (Kohnke et al., 2023; Teng & Shen, 2026).

 

Teachers, who must translate policy and technology into classroom practice, have received less attention than students (Teng & Wu, 2021). Research from Hong Kong illustrates the gap. Chan (2023) surveyed 180 teachers and 457 students to propose an AI policy framework with pedagogical, governance, and operational dimensions. English language teacher educators expected GenAI to reshape curriculum, instruction, and assessment, yet most felt they lacked the confidence and competence to respond (Moorhouse & Kohnke, 2024; see also Moorhouse, 2024). Explicit training can help: an 11-week course raised pre-service language teachers’ professional GenAI competence (Moorhouse et al., 2024). Another recent study by Wu and Miller (2025) offers interesting insights by comparing two university teachers from Macau. The study highlights how distinct approaches to GenAI in teaching are shaped by personal, contextual, and professional factors. Despite this study, comparable evidence from Macau remains scarce.

 

Hong Kong and Macau offer an instructive comparison. Both are highly developed Special Administrative Regions that position themselves as hubs of knowledge and technology, yet their higher education systems differ in scale, institutional culture, and responsiveness to educational change. Comparing teachers in the two cities allows us to examine how similar technological pressures are taken up differently depending on context.

 

Teacher agency in the era of GenAI

The study was framed through the concept of teacher agency. Teacher agency refers to teachers’ capacity to take ownership of their professional growth, encompassing a sense of purpose, competence, autonomy, and reflexivity (Teng, 2019). An ecological perspective holds that agency is not a property that individuals possess but something they achieve through their engagement with particular environments (Biesta & Tedder, 2006; Priestley et al., 2015). In Biesta and Tedder’s (2006) terms, teachers are “actors acting by-means-of-an-environment rather than simply in an environment” (p.19). Agency is also temporal: it draws on past experience, is enacted under present conditions, and is oriented toward imagined futures (Emirbayer & Mische, 1998). This view complements, while shifting emphasis away from, psychological accounts that locate agency primarily in individual self-efficacy and intentionality (Bandura, 2006).

 

Jenkins (2020) distinguishes two forms of teacher agency in response to change. Proactive agency involves teachers initiating change as a personal choice within a supportive environment; reactive agency occurs when teachers are compelled to adapt to changes mandated by leadership or circumstance. This distinction is useful but may be too coarse for a disruptive, fast-moving technology such as GenAI, where institutional mandates, peer cultures, and student behavior change simultaneously.

 

Recent studies support a more dynamic view. Pre-service English teachers took an agentive role in teacher-GenAI collaboration when designing lessons (Lee et al., 2025). University teachers in the United Kingdom enacted agency as a collective response to students’ extensive use of large language models, sometimes marked by anxiety (Kahn et al., 2025). A scoping review of 42 studies on tertiary educators’ agency also notes that earlier reviews largely overlooked university teachers (Cong-Lem, 2024). Few studies, however, compare in-service tertiary teachers across contrasting institutional ecologies.

 

The present study

To address this gap, the present study explores how in-service tertiary teachers in Hong Kong and Macau integrate GenAI into their teaching and how their agency is shaped by context. It is guided by two research questions:

1.      What practices and challenges do in-service tertiary teachers in Hong Kong and Macau report in integrating GenAI into teaching?

2.     How is teacher agency enacted in these two contexts, and to what extent does the proactive–reactive distinction capture it?

 

Methods

Design

The study adopted a qualitative, interpretative design suited to understanding how teachers make sense of GenAI within their own professional settings. Because the project is ongoing, this report uses an instrumental case approach, presenting two information-rich cases selected to illustrate contrasting institutional conditions.

 

Participants

Twenty in-service tertiary teachers took part: 10 from universities in Hong Kong and 10 from universities in Macau, representing a range of disciplines. The two focal participants, Chan (Hong Kong) and Lau (Macau), both pseudonyms, each had approximately ten years of experience in educational technology. Table 1 summarizes the sample.

 

Table 1. Participants

 

Hong Kong

Macau

Participants (n)

10

10

Disciplines

Various

Various

Focal case

Chan

Lau

Focal case experience in educational technology

c. 10 years

c. 10 years

 

 

Data sources and procedure

Three sources of data were collected from each participant: (a) semi-structured interviews exploring participants’ experiences, practices, and perceived challenges in using GenAI; (b) lesson plans illustrating how GenAI was incorporated into teaching and assessment; and (c) written responses to open-ended questions about institutional support, collegial interaction, and future intentions.

 

Analysis

The analytic approach, e.g., reflexive thematic analysis (Braun & Clarke, 2006), with coding informed by the ecological and proactive–reactive frameworks, was conducted. Interview data were triangulated with lesson plans and open-ended responses to compare what teachers said about GenAI with how they incorporated it in planned teaching.

 

Results

The two focal cases revealed sharply different institutional environments but convergent outcomes: both teachers integrated GenAI into their lesson plans. Table 2 compares the cases across the ecological dimensions identified in the data.

 

Table 2. Comparison of the two focal cases

Dimension

Chan (Hong Kong)

Lau (Macau)

Institutional policy

Detailed university guidelines issued swiftly after GenAI emerged

Few institutional or faculty-level AI regulations

Infrastructure and training

Access to multiple technologies; frequent departmental training and workshops

Minimal guidance or infrastructure; no training offered

Collegial culture

Frequent discussion and sharing of ideas, practices, and resources; cross-university seminars

Little communication; colleagues reluctant to adopt GenAI and staying in comfort zones

Main driver

Peer pressure and a competitive, international environment

Personal interest and aspiration to be future-oriented

GenAI in lesson plans

Extensive integration in teaching and assessment

Examples of GenAI use incorporated

Main concern

GenAI overshadowing learning; pace of change

Lack of support and institutional inertia

Self-description

“Forced to go with the flow”

“Dance with shackles on”

Predominant agency

Reactive, with proactive elements

Proactive, with reactive elements

 

 

Chan: “I am forced to go with the flow”

Chan expressed mixed feelings about GenAI’s impact on teaching. His university responded swiftly to its emergence, issuing detailed guidelines and providing access to a range of technologies, and his department frequently organized training sessions and workshops. Beyond this formal support, Chan and his colleagues regularly discussed GenAI, exchanging ideas, practices, and online resources, and he attended seminars hosted by his own and other Hong Kong universities.

 

This supportive ecology was also a competitive one. Chan explained, “You feel the peer pressure in Hong Kong. I don’t want to be left behind.” His lesson plans showed extensive integration of GenAI in both teaching and assessment, which he linked to the city’s professional expectations: “Teachers in Hong Kong must remain cutting-edge and international. We must adapt; otherwise, we risk becoming outdated.”

 

Despite feeling well supported, Chan worried that GenAI might overshadow the learning process. He also noted that students’ growing use of GenAI forced teachers to adapt their practices to AI-generated assignments. He summed up his experience: “I am forced to go with the flow. I wish I could slow down and take a break.”

 

Lau: “I tried to learn to dance with shackles on”

Lau faced far greater obstacles. Although eager to innovate pedagogically through GenAI, she had few opportunities to update her knowledge and practice. Her institution had no AI regulations at either institutional or faculty level, which she described bluntly: “Here we act like ostriches, burying our heads in the sand. I don’t know whether this is due to disinterest or cowardice.” Guidance and infrastructure were minimal, and no training was offered to build teachers’ professional knowledge of GenAI.

 

Lau also reported little communication among colleagues, many of whom were reluctant to adopt GenAI and preferred to remain in their comfort zones. She speculated that Macau’s casino-driven prosperity might have weakened the motivation to seek change.

 

Nevertheless, Lau’s lesson plans included examples of GenAI use. Motivated by personal interest and the aspiration to be a future-oriented teacher, she sought out online resources, attended seminars, and experimented with AI tools and pedagogical approaches on her own initiative. She concluded: “Though my hands are super tied, I tried to learn to dance with shackles on.”

 

Discussion and conclusion

Theoretical implications

The cases of Chan and Lau show that teacher agency in GenAI integration is dynamic and multifaceted, and cannot be fully captured by Jenkins’ (2020) binary of proactive and reactive

agency. Agency emerged instead as a temporal and relational phenomenon, shaped by past experience, enacted within present conditions, and oriented toward future goals (Biesta & Tedder, 2006; Emirbayer & Mische, 1998; Teng, 2019). This echoes recent work treating teachers’ agency with GenAI as negotiated and responsive rather than fixed (Kahn et al., 2025; Lee et al., 2025), and extends it by showing how two very different institutional ecologies produce similar outcomes through different routes.

 

Chan primarily demonstrated reactive agency, responding to a supportive but demanding environment of guidelines, collegial collaboration, and technological infrastructure. Yet his determination to avoid becoming an “outdated” teacher also reflects a proactive, future-oriented element. Institutional and peer pressure to remain competitive and “cutting-edged” compelled him to embrace GenAI, while his wish to “slow down” suggests that a rich environment can constrain as well as enable, leaving teachers little space for reflexivity.

 

Lau, by contrast, exhibited primarily proactive agency, driven by intrinsic motivation and interest in pedagogical innovation. However, her adaptation also involved a reactive dimension: in the absence of institutional guidance and collegial support, she had to respond to these gaps by constructing her own learning pathways. Her metaphor of dancing in shackles captures agency achieved against, rather than through, the environment.

 

The movement between proactive and reactive dimensions in both cases underscores the temporal nature of agency. We therefore argue that teacher agency in the GenAI era is enacted at the intersection of individual intentions and environmental affordances, where teachers navigate tensions between external demands and personal aspirations, and between past, present, and future.

 

Practical implications

The findings suggest that institutional support is necessary but not sufficient. Lau’s context lacked what Yang et al. (2025) calls the operational dimension of GenAI policy: infrastructure and training. Clear GenAI policies, accessible training, and structures for collegial exchange would lower the cost of individual initiative and spread innovation beyond self-motivated teachers; intervention evidence suggests such training can build teachers’ GenAI competence (Moorhouse et al., 2024). In contexts like Chan’s, the challenge is different: institutions may need to temper the pace of change, protect time for reflection, and frame GenAI adoption around learning goals rather than competitiveness. In both settings, teacher education and professional development should cultivate teachers’ reflexive judgement about when and why to use GenAI, not only their technical competence (Moorhouse & Kohnke, 2024).

 

Limitations and future research

This report presents preliminary findings from two of twenty participants, selected for their contrasting contexts and their substantial experience in educational technology. Their views may not represent less experienced or less technology-oriented colleagues, and the findings are not intended to be generalized to each city as a whole. The data are also largely self-reported. The full analysis will examine patterns across all twenty participants, compare disciplines, and trace how agency develops over time as institutional GenAI policies mature.

 

Conclusion

As GenAI reshapes higher education, teachers are neither passive recipients of change nor fully autonomous innovators. Chan’s going with the flow and Lau’s dancing in shackles illustrate two ways agency is achieved under very different conditions. Recognizing agency as temporal and relational can help institutions design GenAI policies and professional development that support teachers’ purpose, competence, autonomy, and reflexivity.

 

 

References

Bandura, A. (2006). Toward a psychology of human agency. Perspectives on Psychological Science, 1(2), 164–180. https://doi.org/10.1111/j.1745-6916.2006.00011.x

Biesta, G., & Tedder, M. (2006). How is agency possible? Towards an ecological understanding of agency-as-achievement. Learning lives: Learning, identity, and agency in the life course. Working paper five, Exeter: Teaching and Learning Research Programme.

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa

Chan, C. K. Y. (2023). A comprehensive AI policy education framework for university teaching and learning. International Journal of Educational Technology in Higher Education, 20, 38. https://doi.org/10.1186/s41239-023-00408-3

Cong-Lem, N. (2024). Teacher agency for change and development in higher education: A scoping literature review. International Journal of Educational Reform, 35(3),992-1015. https://doi.org/10.1177/10567879231224744

Emirbayer, M., & Mische, A. (1998). What is agency? American Journal of Sociology, 103(4), 962–1023.

Jenkins, G. (2020). Teacher agency: The effects of active and passive responses to curriculum change. The Australian Educational Researcher, 47(1), 167–181. https://doi.org/10.1007/s13384-019-00334-2

Kahn, P., Carrigan, M., Smith, P., Murtagh, L., Liu, R., & Song, F. (2025). Teacher agency and generative artificial intelligence: Teaching in higher education as a responsive, cultural activity. Learning, Media and Technology. Advance online publication. https://doi.org/10.1080/17439884.2025.2575993

Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., Krusche, S., Kutyniok, G., Michaeli, T., Nerdel, C., Pfeffer, J., Poquet, O., Sailer, M., Schmidt, A., Seidel, T., … Kasneci, G. (2023). ChatGPT for good? On opportunities and challenges of large language models for education. Learning and Individual Differences, 103, 102274. https://doi.org/10.1016/j.lindif.2023.102274

Kohnke, L., Moorhouse, B. L., & Zou, D. (2023). ChatGPT for language teaching and learning. RELC Journal, 54(2), 537–550. https://doi.org/10.1177/00336882231162868

Lee, S., Jeon, J., & Choe, H. (2025). Generative AI (GenAI) and pre-service teacher agency in ELT. ELT Journal, 79(2), 287–296. https://doi.org/10.1093/elt/ccaf005

Moorhouse, B. L. (2024). Beginning and first-year language teachers' readiness for the generative AI age. Computers and Education: Artificial Intelligence, 6, 100201. https://doi.org/10.1016/j.caeai.2024.100201

Moorhouse, B. L., & Kohnke, L. (2024). The effects of generative AI on initial language teacher education: The perceptions of teacher educators. System, 122, 103290. https://doi.org/10.1016/j.system.2024.103290

Moorhouse, B. L., Wan, Y., Wu, C., Kohnke, L., Ho, T. Y., & Kwong, T. (2024). Developing language teachers’ professional generative AI competence: An intervention study in an initial language teacher education course. System, 125, 103399. https://doi.org/10.1016/j.system.2024.103399

Moorhouse, B. L., Yeo, M. A., & Wan, Y. (2023). Generative AI tools and assessment: Guidelines of the world's top-ranking universities. Computers and Education Open, 5, 100151. https://doi.org/10.1016/j.caeo.2023.100151

Priestley, M., Biesta, G., & Robinson, S. (2015). Teacher agency: An ecological approach. Bloomsbury.

Teng, M. F. (2019). Autonomy, agency, and identity in teaching and learning English as a foreign language. Springer.

Teng, M. F., & Wu, J. G. (2021). Tea or tears: online teaching during the COVID-19 pandemic. Journal of education for teaching, 47(2), 290-292. https://doi.org/10.1080/02607476.2021.1886834

Teng, M. F., & Shen, X. (2026). Digital applied linguistics: What is it and where are we in the GenAI era? (2026). Digital Applied Linguistics, 4, 103594. https://doi.org/10.29140/dal.v4.103594

Watson Todd, R. (2026). Using generative AI to revise articles: How much is too much? Applied Language Sciences, 2(1). https://doi.org/10.65553/ALS.260108

Wu, J. G., & Miller, L. (2025). Smart or sweat: The bittersweet journey of teachers’ AI literacy. International Journal of TESOL Studies, 1, 1-10. https://doi.org/10.58304/ijts.250124

Yang, Z., Wu, J. G., & Xie, H. (2025). Taming Frankenstein’s monster: Ethical considerations relating to generative artificial intelligence in education. Asia Pacific Journal of Education, 45(4), 1330-1343. https://doi.org/10.1080/02188791.2023.2300137

 

 

 

 

 

Author bio

 

Junjie Gavin Wu is an Assistant Professor and Ph.D. Supervisor at Macao Polytechnic University, as well as the Vice President of PacCALL and the Vice President of GloDAL. He is Editor-in-Chief of SN Social Sciences (Scopus, Springer), Editor-in-Chief of Applied Language Sciences, an Executive Associate Editor of Computers & Education: X Reality (ESCI), and an Executive Associate Editor of Artificial Intelligence in Language Education (Castledown). Gavin has authored around 100 English publications, with over 40 papers appearing in SSCI/SCI journals. He has published books with Cambridge University Press, Routledge, and Springer, and spearheaded over 10 special issues. Gavin serves on the editorial boards of several leading journals, such as the British Journal of Educational Technology, Interactive Learning Environments, Language Learning & Technology, ReCALL, Computers and Education: Artificial Intelligence, IJCALLT, Multimedia-Assisted Language Learning, and so on. He is also an award-winning teacher in China.

E-mail: gavinjunjiewu@gmail.com

ORCID: 0000-0003-4937-4401