.JPG)
.JPG)
.JPG)
Many talk about AI as though it is another initiative or an addition to existing school priorities, something to add to the school development plan (See Pic 1), but it shouldn’t necessary be viewed like that.
AI should be viewed more as a method to help achieve your school priorities rather than just another priority. (See Pic 2).
Thinking like this moves us away from treating AI as a standalone flashy, showcase tool, towards integrating it as a subtle yet powerful support assistant.
Visitors to schools where AI is embedded might not even see the AI at work in the school. Good AI integration in schools should be invisible:
Less about showcase displays and more about the subtle, day to day support it provides.
Not about creating a wow factor with a bot in the classroom, it's about using AI to assist with the mundane tasks that consume valuable time or support better learning opportunities for students.
It's not about a preset library of prompts, but rather bespoke individual prompts tailored to the unique needs of the person, the individual task and sometimes the tool itself.
It's not about one big assembly on AI safety or a few explicit AI lessons, it's about constant conversations around AI nutrition.
Embedding AI use is about getting down and dirty, weaving it into the fabric of the school.
A suggested training process works something like this:
1. Problem or challenge identified by school.
2. Current method analysed and broken into steps.
3. LLM demo of how AI can speed up and enhance each step.
4. After each demo, staff go hands on, experimenting to create their own solutions and push LLM further.
5. Sharing and discussion after each step.
6. Comparison of pre-AI vs. post AI final output and process.
This may mean that your training is more impactful than training generically on a theme such as lesson planning or introducing a particular tool. It also builds agency and ensures staff are invested in their output.
Some may advocate for centralised control and generic implementation of AI in schools, but true transformation happens at the granular level. To create effective strategy and policies, we must first experiment and understand the intricacies of how AI can be beneficial, risky, or simply not useful in specific contexts. No one has done this before so I think we need a balance of top down underpinned by a bottom up approach.
Pic 3 is from a recent survey I did of subject leaders across Surrey schools when asked “Rank these in order of most to least influence on your ability to embed AI into your teaching practice” it’s no surprise what they need or value.
Obviously, policy, safety and structure are hugely important and should not be ignored but it should also be underpinned by hands on experience and sharing. Staff may say they don’t need celebration and recognition but it’s vital to add momentum and motivation.
Prioritise change, and the impossible becomes achievable.
Key Learning
AI should be integrated into schools as a method to achieve existing priorities, not as an additional priority.
Effective AI integration is subtle and supports day-to-day tasks, rather than being a flashy, standalone tool.
A balance of top-down and bottom-up approaches is crucial for effective AI strategy and policy in schools.
Risks
Treating AI as a standalone tool or an additional priority could lead to ineffective integration and underutilisation.
Overemphasis on centralised control and generic implementation may hinder the transformative potential of AI at the granular level.
Lack of hands-on experience and sharing could limit the ability of staff to effectively embed AI into their teaching practice.
