5 Levels of Teacher AI Literacy: From Awareness to Architect

When I run CPD sessions with schools, I always start with the same question: "On a scale of 1 to 5, how AI literate do you think you are?" The responses are predictable. Most teachers put themselves at a 2 or 3. A few confident ones say 4. Nobody says 5. And nobody can tell me what any of those numbers actually mean.

That is the problem. We ask teachers to rate their AI literacy without giving them a framework for what the levels look like. It is like asking someone to rate their fitness without telling them whether 1 means "cannot walk to the shops" or "runs a 5K slowly." Without defined levels, self-assessment is meaningless; and CPD built on meaningless assessment is wasted investment.

This article presents a five-level progression model for teacher AI literacy. It is drawn from my analysis of 33 international frameworks, particularly UNESCO's AI Competency Framework for Teachers, and refined through practical application with schools across the Middle East, the UK, and internationally. It is designed to be used, not just read: use it to baseline your staff, plan your CPD, and measure progress.

Level 1: Aware

What it looks like: The teacher knows that AI tools exist in education. They have heard of ChatGPT, possibly Copilot or Gemini. They may have tried one once out of curiosity. They understand at a very high level that AI generates content, but they could not explain how. They have opinions about AI, usually either "it is going to replace us" or "it is just a fad", but these opinions are not informed by structured understanding.

What it does not look like: The teacher is not using AI in their professional practice in any regular way. They may be actively avoiding it. They are unlikely to have engaged with any AI-related CPD beyond a staff briefing.

What they need: Demystification. Level 1 teachers need to understand what AI actually is: not the science fiction version, but the practical reality of pattern recognition, language models, and machine learning at a conceptual level. They need permission to experiment without fear of getting it wrong. And they need to see practical examples of how teachers at their career stage, in their subject, are using AI successfully.

Level 2: Explorer

What it looks like: The teacher regularly uses AI tools in their professional practice: generating lesson resources, drafting emails, creating differentiated materials, brainstorming ideas. They have a go-to tool (usually ChatGPT or a subject-specific platform) and have developed basic prompt strategies through trial and error. They can identify when AI output is useful and when it is not, but they may not be able to articulate why.

What it does not look like: The teacher is not yet thinking critically about the AI tools they use. They are not considering data protection implications, bias in outputs, or the pedagogical rationale for AI integration. They are using AI as a productivity tool, not as a teaching tool.

What they need: Structure. Level 2 teachers benefit from understanding prompt engineering principles: not as a technical skill but as a communication skill. They need exposure to the school's AI policy so they understand the governance context for their experimentation. And they need guided reflection; not "did the AI tool help?" but "what did the AI tool add to student learning that was not there before?"

Level 3: Practitioner

What it looks like: The teacher integrates AI into their teaching practice with intention and purpose. They design learning activities that use AI as a pedagogical tool: not just for content generation, but for student learning. They might use AI to create Socratic questioning sequences, to generate worked examples at different difficulty levels, or to provide students with AI-generated texts to critique and improve.

They understand the school's AI policy and follow it consistently. They can explain to students when AI use is appropriate and when it is not, and they model responsible AI use in their own practice. They have an awareness of data protection principles; they know not to put student names into public AI tools, and they understand why.

What it does not look like: The teacher is not yet evaluating AI tools against formal criteria. They use what works but may not have a framework for assessing whether a tool is pedagogically sound, ethically acceptable, or compliant with data protection requirements. They are skilled users, but they are not yet critical evaluators.

What they need: Depth. Level 3 teachers are ready for more substantive engagement with the ethical and evaluative dimensions of AI. They benefit from understanding bias in AI systems, exploring how different AI tools handle student data, and learning about the international framework landscape. They are also ready to start contributing to wider school discussions about AI; for example, through department-level planning sessions or contributions to the AI steering group.

Level 4: Evaluator

What it looks like: The teacher can critically evaluate AI tools and their impact on teaching and learning. They assess tools not just on usability or convenience but on pedagogical value, ethical implications, data practices, and alignment with educational objectives. They can identify bias in AI outputs and explain its implications to colleagues and students.

They contribute to school-wide AI governance: reviewing tools before adoption, advising on policy, supporting colleagues' development. They may lead department-level AI integration, designing schemes of work that thoughtfully incorporate AI and modelling best practice for their team.

They engage with the broader conversation about AI in education: reading research, attending conferences, connecting with practitioners in other schools. They have an informed view of where the field is heading and can anticipate how changes in AI capability might affect their practice.

What it does not look like: The teacher is not yet designing CPD for others or contributing to whole-school strategy at a senior level. They are an expert practitioner and a critical evaluator, but they are not yet a leader in this space.

What they need: Platform and influence. Level 4 teachers need opportunities to lead, CPD sessions, steering group membership, cross-school collaboration. They benefit from deeper engagement with the international framework landscape, understanding how different frameworks define competency and how their school's approach fits within that landscape. And they need recognition: these are your AI champions, and they should be valued and supported accordingly.

Level 5: Architect

What it looks like: The teacher operates at the intersection of AI, pedagogy, and institutional strategy. They design and deliver AI CPD programmes for colleagues. They contribute to whole-school AI strategy, working with senior leadership to define the school's AI vision and priorities. They understand the international framework landscape in detail; they can position the school's approach within it.

They may be building or customising AI tools for educational use: not necessarily coding, but designing prompts, workflows, and integration patterns that others in the school can use. They engage with the professional community beyond their school, sharing practice, publishing insights, and contributing to the wider conversation about AI in education.

What it does not look like: Level 5 is not about being a technology enthusiast. It is about combining deep pedagogical expertise with AI knowledge to shape institutional practice. The best Level 5 practitioners I work with are experienced educators first and AI experts second. Their value comes from understanding how AI serves education, not from understanding AI for its own sake.

What they need: Autonomy, resources, and a strategic mandate. Level 5 teachers need time and space to fulfil their role; this cannot be an add-on to a full teaching timetable. They need a budget for professional development, conference attendance, and resource development. And they need a clear mandate from senior leadership that positions AI literacy as a strategic priority, not a side project.

Using the Model

This model is designed to be used in three ways.

Baselining. At the start of the academic year, have every teacher self-assess against the five levels. Supplement with evidence where possible; a teacher who claims Level 3 should be able to show examples of AI-integrated lesson design. Use the aggregate data to understand your school's competency profile: what proportion of staff are at each level?

Planning CPD. Design your CPD programme to meet teachers where they are. A school where most staff are at Level 1 needs a fundamentally different programme from a school where most are at Level 3. Avoid one-size-fits-all training; it bores the advanced practitioners and overwhelms the beginners.

Measuring progress. Re-assess at the end of the year. Has the distribution shifted? Are more teachers at Level 3 and above? Can you demonstrate measurable improvement? This data is gold for governance reporting, accreditation evidence, and strategic planning.

The AI Literacy Audit Tool assesses your school's teacher competency provision against the expectations of 33 international frameworks. It tells you not just where your staff are, but where the global consensus says they should be; and it highlights the gap in concrete, actionable terms.

Whether you use the tool or this five-level model, the point is the same: AI literacy is a progression, not a checkbox. Define the levels, measure against them, and invest in the journey from one to the next. That is how you build a genuinely AI-competent teaching workforce.