Every school needs an AI strategy, but what should it actually achieve?

An AI strategy should not be a declaration of enthusiasm for technology. It should be a plan for improving education while managing risk. Calls for every school to develop an AI strategy are growing louder. The argument is understandable. AI is already affecting teaching, learning, assessment, administration and pupils' lives beyond school. Institutions need a coherent response.

However, there is a danger that the strategy itself becomes the objective. A governing body approves a document. A school appoints an AI champion. Staff attend a training session. A list of approved tools is circulated. All may be useful steps, but none, on its own, demonstrates that the school has become AI-ready. The measure of a successful AI strategy is not how much AI a school uses. It is whether the school is better equipped to improve education and manage the consequences of AI.

Start with the educational purpose

A strategy should begin with the school's priorities, not a catalogue of technological possibilities. Where might AI help teachers spend more time on the work that matters most? Could it support better planning, accessibility or feedback? What do pupils need to understand about AI to become informed, independent learners? Equally, where might AI undermine learning, create unacceptable risks or introduce unnecessary complexity? These are educational and leadership questions that need to be answered by any effective AI strategy.

Five questions for school leaders

A meaningful AI strategy should address five connected areas.

1. Are our staff AI-literate? Staff need a shared understanding of AI's capabilities, limitations and risks, supported by continuing professional development rather than a one-off introduction.

2. Are our pupils being prepared? AI literacy should be developed progressively, including critical thinking, safe use, academic integrity and understanding when independent work matters.

3. Are our policies and safeguards fit for purpose? Schools need clarity on approved tools, personal data, safeguarding, assessment, accountability and human oversight.

4. Are we improving teaching and learning? AI should be evaluated against educational outcomes, inclusion, workload and the quality of pupils' learning and not the number of applications being used.

5. Are governors providing effective oversight? Governing bodies should understand the opportunities and risks well enough to ask informed questions, support leaders and monitor impact.

These questions are interconnected. A school cannot sensibly separate AI training from safeguarding, or classroom use from assessment.

Implementation matters more than publication

The strongest strategies will be living plans. They will identify priorities, establish responsibilities, provide training and create opportunities to evaluate practice. They will be reviewed as technology develops and as evidence emerges. They will also allow professional judgement. A teacher who decides not to use AI for a particular task may be making an entirely sound educational decision. A school that proceeds carefully in a high-risk area should not be regarded as lacking ambition. Preparedness is not the same as indiscriminate adoption.

From strategy to confidence

The AiEdCertified Framework was developed to help schools approach this work systematically, considering AI literacy, policies and ethics, tools and systems, digital pedagogy, and collaboration. Its purpose is not to prescribe a single technological future. It is to help institutions make informed decisions that reflect their educational responsibilities.

Every school should be able to explain what it wants AI to achieve, how it will protect pupils and staff, and how it will know whether its approach is working. Government should expect every school to be AI-ready, not require every school to be AI-heavy. That is the distinction on which a successful national approach may depend.

‍