Featured image source: Catherine Breslin & Tania Duarte / https://betterimagesofai.org / https://creativecommons.org/licenses/by/4.0/
RACHELE NEWMAN, DIRECTOR OF TEACHER PROFESSIONAL LEARNING, UNIVERSITY OF SOUTHAMPTON, UK
SIAN CUMMING, DEPUTY HEADTEACHER, UPPER SHIRLEY HIGH SCHOOL, UK
NATALIE BUNGAY, DEPUTY HEADTEACHER, WESTON SECONDARY SCHOOL, UK
EMILY BELL, CO-HEADTEACHER, REDBRIDGE COMMUNITY SCHOOL, UK
The benefits for teachers in engaging with research-rich professional activity include improved subject and pedagogical knowledge, as well as enhanced skills in adaption and confidence (Cordingly, 2015; Timperley et al., 2007). Using research evidence to support practice requires careful engagement with the research outcomes. Brown et al.’s (2022) model of research-informed educational practice (RIEP) outlines the steps involved in teachers engaging in effective RIEP, which include accessing, critiquing and applying education research to their practice. The rise of AI presents new possibilities to support teachers in this process by reducing some of the documented barriers to RIEP: locating and accessing high-quality research (Cordingly, 2015) and navigating the complex and lengthy format of academic papers (Cain, 2019). Recent research (Sowa et al., 2025) has revealed that teachers are using AI in a variety of ways to support their RIEP. Generative AI can summarise, simplify and contextualise research evidence, making it more actionable for teachers. Further findings indicate that teachers are using AI for lesson ideation, with generative AI being used to support creative thinking and to personalise content. Some teachers also actively refine prompts, seek peer input and critically evaluate AI outputs to overcome current limitations, indicating a growing awareness of the need for cautious and context-sensitive use of AI (Sowa et al., 2025).
However, this study revealed questions about what this level of engagement with AI means for teacher professional identity and agency. This aligns with wider national discourse exploring a decline in teacher autonomy and agency in England in recent years (Newman, 2025; Masson, 2025). A report commissioned by the National Education Union, reviewing the widescale use of centralised curriculum materials, has further pointed to the erosion of teacher agency and autonomy resulting from teaching materials that they have not conceptualised, created or designed (Traianou et al., 2025). It is possible that a similar reduction of agency could be experienced by teachers who work with AI-generated teaching materials.
So, if AI is being used by teachers to support their knowledge development, pedagogical decisions and content creation – traditionally the unique, core intellectual domain of the teacher – then what effect might this be having on their professional identity? Teacher identity is recognised as dynamic and flexible (Rushton and Reiss, 2021). As such, it could be expected that professional identity might flex in response to the new ways of working that AI can offer. Some teachers report that their professional identity is compromised by overreliance on AI, struggling to distinguish their own contributions from AI-generated content. Others use AI to validate their existing beliefs, highlighting both empowering and challenging effects on their professional identity.
A recent UNESCO position paper advocates for the centrality of teacher agency in the age of AI (Teacher Task Force, 2025). Key points raised include how to prevent AI from weakening critical thinking and creativity, how teachers can leverage AI to expand teacher autonomy and professional judgement, and the importance of AI as a complement and not a substitute for pedagogical expertise. These points, coupled with the findings by Sowa et al. (2025), suggest a possible need to reconceptualise key notions relating to a teacher’s identity and their role, what a teacher is and what knowledge they actually require. What is the role of a teacher if they are not required to conceive of or create original lesson materials? And what impact might it have on teacher subject, disciplinary and pedagogical knowledge when lessons and teaching materials are co-authored with AI?
The research
What follows is a summary of the pilot phase of a research project conducted in collaboration between a university-based teacher educator in the South of England and senior leaders in partner schools. The project aimed to explore teacher experiences of using AI to support their practice and to gain their thoughts on any impact on their professional identity and autonomy. This pilot phase involved a convenience sample of three secondary schools within the same city, where the senior leaders had indicated a desire to collaborate in the project and therefore acted as gatekeeper to participants. A mixture of quantitative and qualitative data was collected via an anonymous survey tool, with 45 separate participant responses received. There was representation across core, foundation (humanities and religions studies), creative arts and physical education subjects, and participants reported teaching experience of between one and 10+ years.
How is AI being used?
Table 1 summarises a range of ways in which respondents are using AI in their work as a teacher.
Table 1: Ways in which AI is used in teaching
| AI is used to: | |||
| support some aspect of my practice | 98% | ||
| support administrative tasks (emails/letters/data analysis/report writing) | 56% | ||
| create materials for students to use in lessons (e.g. worksheets) | 60% | ||
| adapt resources for students | 40% | ||
| generate questions to ask of students | 24% | ||
| support my marking, assessment and feedback | 27% | ||
| supplement my subject knowledge | 23% | ||
| support creative lesson ideas | 8% | ||
| support my lesson planning | 11% | ||
Almost every respondent is using generative AI models in a range of ways to support their practice, including for administrative or bureaucratic tasks. Data indicates widespread use of AI to support with the planning and preparation/creation of teaching-related materials. The most commonly reported usage was for creating student-facing lesson materials (e.g. worksheets), but other significant examples included the use of AI to adapt resources to meet specific needs of students. The preparation of lesson resources has historically been part of the time-consuming but lower-level work associated with lesson preparation. These materials are generated after the core, deep thinking about a lesson has happened, and so arguably require less cognitive ownership from a teacher, making these possibly more appropriate for outsourcing to AI generation.
However, some participants report using AI to engage in activity that could be said to be closer to the direct knowledge related to teaching and learning, or engaging in the deeper thinking that otherwise would have been the purvey of the teacher. Twenty-seven per cent of respondents report using AI to support marking, assessment and feedback. Twenty-three per cent use AI to supplement their knowledge, including subject knowledge (four per cent), suggesting creative ideas for lesson activities (eight per cent), generating questions for teachers to ask (24 per cent) and producing lesson plans (11 per cent) – although it is difficult to know from this how much of the teacher thinking has happened and how far AI is simply consolidating this into a coherent lesson outline.
Professional identity
This research reveals varied responses in relation to the impact of AI on teacher professional identity. Twenty-four per cent of respondents report that AI supports their day-to-day work as a teacher, acknowledging that it is simply another tool to be leveraged. Perceiving AI in this way suggests that these teachers perhaps don’t regard as AI having any more effect on their professional identity than other commonly used classroom tools. In fact, 20 per cent of respondents report AI as having no impact on their professional identity, with comments such as ‘it doesn’t affect my professional identity; it is just a tool I use at times when time is short’ (T1) being indicative of a range of responses.
Fourteen per cent of teachers report that their professional identity has been enhanced by their use of AI. For some, AI serves to underscore the importance of the relational aspects of teaching and provides a recognition that this is (currently) something that AI cannot replicate or replace. This reinforces the human element of teaching and serves as a reminder of the importance of this for teachers’ professional identity. Furthermore, AI is reported to be carrying the workload burden of teachers, thereby freeing them up to ‘work on the creative and expressive side of teaching’ (T2). One interesting reflection provides food for thought in terms of how teachers may be able to use AI as support rather than replacement: ‘AI is reshaping my identity as a teacher more deeply than expected; it helps me guide, design and mentor with greater focus, creativity and connection.’ (T3) This suggests a shift in the focus of teachers’ work, away from the time-consuming ‘heavy lifting’ work traditionally done by teachers (generating/adapting/sourcing teaching materials, administrative tasks, etc.) towards a new focus on ideas, relationships and creativity, which arguably have been sidelined in recent years due to excessive workload and bureaucracy.
However, in line with Sowa et al.’s study (2025), seven per cent of teachers suggested that their professional identity is depleted by their use of AI, with individuals indicating that they ‘worry about lack of ownership’ (T4) and that ‘things are going to be overtaken’ (T5). These comments might belie a lack of confidence and experience of AI or might indicate deeper concerns relating to the erosion of the intellectual process central to a teacher’s work. These concerns have been mirrored in a variety of comments, with one teacher noting that the ‘quality of AI output is related to the quality of input; this quality comes from the deep level of subject knowledge and I worry this could be lost’ (T6). In order for teachers to generate effective prompts to produce suitable AI content, it is important that teachers continue to maintain their high levels of subject, disciplinary and pedagogical knowledge, and AI should not be ‘a replacement for knowledge, reading or planning’ (T7). Emphasised here is the importance of teachers being able to exercise their own professional judgement and agency in relation to their engagement with AI, how they use it and to what level they relinquish ownership of the core intellectual elements of their role as a teacher. Professional judgement and agency are intricately linked to teacher identity (Masson, 2025), and so it is crucial that teachers retain agency and autonomy over their professional decisions in relation to their practice and how AI might interact with this.
I would suggest that we are on the cusp of a fundamental reconceptualisation of the role of a teacher in the digital world. Teachers are already making widespread use of AI tools to support their practice, and some clearly consider the time-saving benefits of AI to be a worthy trade-off for any possible compromise to professional identity. The use of AI could be a real game-changer in a profession where the excessive workload of teachers is so well documented. But we must pay close attention to the work that AI is actually doing and protect the professional integrity, agency, autonomy and knowledge of the teachers who are creating and shaping the educative experiences of our young people. As a profession, we need to think clearly about what we define the core work of a teacher to be, and must engage in greater conversation to support teachers to consider what it means for our professional identity when we co-author/co-create learning experiences for our students.
The examples of AI use and specific tools in this article are for context only. They do not imply endorsement or recommendation of any particular tool or approach by the Department for Education or the Chartered College of Teaching and any views stated are those of the individual. Any use of AI also needs to be carefully planned, and what is appropriate in one setting may not be elsewhere. You should always follow the DfE’s Generative AI In Education policy position and product safety standards in addition to aligning any AI use with the DfE’s latest Keeping Children Safe in Education guidance. You can also find teacher and leader toolkits on gov.uk.
