JAMES CLEMENTS, HEADMASTER, FOREST PREPARATORY SCHOOL, UK
Generative artificial intelligence (GenAI) is no longer a peripheral concern for schools. Teachers and pupils already encounter chatbots, image generators and lesson-planning assistants in everyday work. In England, the Department for Education acknowledges that GenAI offers ‘exciting opportunities’ to improve education by helping teachers to focus on teaching and reducing administrative burdens (DfE, 2025). The guidance stresses that AI can support every child if it is used safely and effectively, with the right infrastructure and ongoing development of evidence (DfE, 2025). At the same time, Ofsted’s early-adopter research warns that doing nothing about AI may itself be risky, yet any benefits depend on clear oversight, training and governance (Ofsted, 2025).
These statements illustrate the leadership challenge: GenAI will not improve schooling by default. While leaders cannot control whether pupils encounter AI outside school, they can control why and how it enters professional practice. Schools can adopt AI through deliberate design, shared values and ethical scrutiny, or allow it to drift in through experimentation and convenience. The question is not simply whether AI will be used, but how leaders shape its purposes and limits.
Evidence suggests that GenAI can, when used intentionally, amplify creativity and efficiency. A systematic review of 19 experimental studies on AI-enhanced creative pedagogy found that AI interventions produced large, statistically significant improvements in creative thinking. Fluency improved most, while gains in originality and elaboration required structured scaffolding and reflective use (Alshehri et al., 2026). The review also noted concerns about cognitive dependency and authenticity, emphasising that AI works best as a reflective partner rather than an autonomous replacement for human creativity (Alshehri et al., 2026).
Generative tools can provide rich creative stimuli. Large language models such as ChatGPT can generate alternative perspectives, stylistic variations and conceptual prompts that provoke curiosity and imaginative divergence.
Gill-Simmen (2025) characterises this potential as a ‘pedagogy of wonder’. When teachers frame AI as an exploratory partner and encourage pupils to critique and adapt its outputs, they invite questioning rather than rote reproduction.
For teachers, the most immediate benefit is time. A randomised controlled trial funded by the Education Endowment Foundation showed that teachers using ChatGPT alongside a structured guide reduced their Year 7 and 8 lesson-planning time by an average of 25 minutes per week, bringing planning time down from 81.5 to 56.2 minutes without reducing quality (EEF, 2024). Freed from routine planning, teachers can reinvest time in formative feedback, adaptive questioning and creative sequencing – the work that is hardest to automate.
Yet as the Ada Lovelace Institute and Nuffield Foundation caution, evidence of impact remains uneven; access, governance and implementation quality are decisive factors. Romero’s systematic review of AI-enhanced creative engagement found that most empirical studies occur at low levels of interaction, interactive consumption and individual content creation; few involve collaborative creation and none reach participatory co-creation. Teachers primarily use AI at these lower levels, and the review calls for professional development and AI literacy to move towards higher-order creative engagement (Romero, 2025).
Alongside these opportunities sit significant risks. GenAI can erode human agency if treated as authoritative rather than provisional. A UK mixed-methods study on teacher agency and GenAI argued that the challenges of AI are best framed as a relationship between teacher and student agency: generative tools can shift epistemic agency away from educators and learners and therefore require dialogue and collective response in order to preserve agency (Kahn et al., 2025). The study emphasised that teacher agency is not fixed but emerges from cultural and pedagogical activity, and that educators must negotiate the uncertainty introduced by large language models (Kahn et al., 2025).
Dependency is a real concern. The systematic review mentioned previously warned that improvements in fluency may be offset by cognitive dependency and authenticity erosion if AI becomes a substitute for thinking (Alshehri et al., 2026). Pupils are aware of this risk: a recent survey of 2,000 UK secondary pupils found that 60 per cent worry that AI encourages copying rather than originality and 48 per cent want more guidance from teachers on responsible AI use (Oxford University Press, 2025). Meanwhile, the DfE’s guidance acknowledges that evidence is still emerging on the benefits and risks of pupils using GenAI, and commits to considering risks alongside opportunities, ensuring safety and reliability and addressing barriers to effective use (DfE, 2025).
Accountability pressures can amplify these risks. When AI tools prescribe curriculum content or assessment criteria with limited scope for teacher interpretation, professional autonomy diminishes. If GenAI becomes the arbiter of quality, then originality and agency narrow; if it becomes a stimulus for critique and refinement, then agency can grow. Human oversight is therefore essential. The European Data Protection Supervisor (2025) warns that automated decision-making introduces risks of opacity, bias and discrimination, and that rigorous monitoring and meaningful human involvement are necessary in order to safeguard fundamental rights.
Human-centred leadership as a strategic response
If GenAI is to enhance rather than diminish creativity and agency, then leadership must be intentional, ethical and pedagogically grounded. The DfE guidance (2025) makes clear that responsibility for accuracy and appropriateness remains with teachers and schools, even when AI tools are used. Professional bodies such as the National Association of Head Teachers stress that AI cannot replace human expertise and that leaders require clear policies, training and ethical frameworks. Four priorities follow:
- Vision and values: Leaders must articulate a clear educational rationale for AI use. Without this, adoption becomes reactive, driven by workload pressure rather than purpose. The key question is: does this use of AI expand pupils’ capacity to think, create and decide, or does it simply increase the speed of production?
- Professional trust and development: Teacher agency depends on trust and competence. Professional development should address not only how generative tools function, but also how they can be used critically, ethically and creatively. The EEF trial (2024) shows that guidance matters as much as access. Romero’s review (2025) similarly calls for training to move from low-level consumption to participatory co-creation.
- Pupil agency and co-construction: If GenAI becomes commonplace, then pupils need explicit teaching in critique, revision and ethical reasoning. Dialogue about AI use should involve pupils, teachers and parents. Students should experience generative tools as partners for testing ideas and strengthening drafts, and not as authorities that replace thought.
- Ethical governance and equity: AI governance is a safeguarding and inclusion issue. Policies must ensure transparency, fairness and human oversight; they must address the digital divide and the potential for AI to exacerbate inequalities. Professional and regulatory frameworks should guide procurement, data protection and consent. The European Data Protection Supervisor (2025) notes that simply inserting a human does not automatically ensure accountability; oversight must be meaningful and supported by training and auditing.
GenAI raises sharp questions for assessment. While adaptive systems can personalise feedback and identify patterns, creativity resists simple measurement. Automated marking tools may struggle with nuance and originality, and they can replicate biases present in training data. Human judgement remains essential for evaluating creative work, providing narrative feedback and ensuring fairness. GenAI can assist with formative feedback and efficiency, but it should not redefine quality. Hybrid assessment models, combining AI-generated insights with professional interpretation, offer a way forward.
GenAI presents schools with a profound leadership challenge. It can expand creativity, strengthen agency and support more inclusive learning. It can also narrow professional judgement and pupil voice if adopted uncritically, or foster dependency if pupils learn to outsource thinking. The determining factor is leadership as discernment: engaging actively, setting values-led boundaries, investing in professional learning and designing human-centred practices, where AI supports judgement rather than replaces it. The evidence is clear: GenAI can yield large gains in creative thinking and reduce teacher workload (Alshehri et al., 2026; EEF, 2024), but its effectiveness depends on pedagogy, scaffolding and ethics. Systematic reviews show that most current practice remains at low levels of creative engagement (Romero, 2025). Teacher–student agency must be consciously protected and co-constructed (Kahn et al., 2025), and regulators warn that human oversight is essential (European Data Protection Supervisor, 2025). Leaders who respond with clarity, courage and care can harness GenAI as a powerful ally to human ingenuity; those who do not may find that AI entrenches the very problems that it promises to solve.
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.
