The global technology industry has settled on a four-layer framework for AI competency. It tells you what every worker needs to know, from the boardroom to the shop floor. And it reveals exactly where schools are failing.
Layer 1 is foundational decision skills. Reading data. Interpreting visualisations. Telling a story with evidence. 85% of enterprise leaders say data-driven decision making is essential for daily operations. This is not coding. This is thinking clearly with information.
Layer 2 is fluency and responsible use. Understanding what AI is, how businesses apply it, what the ethical boundaries are, and how to use tools like copilots effectively. 78% of enterprises say basic AI concepts are now a baseline requirement for all non-technical roles. Not some roles. All of them.
Layer 3 is core technical foundations. Python. SQL. Prompt engineering. Machine learning. Retrieval-augmented generation. This is role-dependent but increasingly relevant for marketing, HR, and product teams, not just engineers.
Layer 4 is advanced development. Neural networks. Transformers. MLOps. Deep learning architecture. This is specialist territory, and it commands the highest salaries. Machine learning engineers start above $110,000.
Where schools fit
Most schools are not teaching any of these layers intentionally. A few Computing departments touch Layer 3 and 4 content. Almost nobody is systematically building Layer 1 and Layer 2 competencies across the school.
That matters because Layer 1 and Layer 2 are the ones every student needs, regardless of what career they end up in. A nurse needs Layer 1 skills to interpret AI-assisted diagnostic data. An architect needs Layer 2 understanding to work responsibly with AI design tools. A journalist needs both to evaluate AI-generated content and explain algorithmic bias to readers.
These are not specialist skills. They are the new baseline. And the data says 88% of enterprise leaders consider basic data literacy essential, while 72% require baseline AI fluency. Those numbers describe the workforce your students are heading into.
The certification explosion
The gap between what employers need and what education provides has created an explosion in professional certifications. Google, AWS, and Microsoft all offer technical AI certifications for engineers. But the fastest-growing segment is vendor-neutral, business-focused credentials like the Certified AI Practitioner and CertNexus AI Biz, designed for managers and non-technical professionals.
15% of all marketing job postings now explicitly require AI competency. 9% of HR postings do too. This is not a tech sector issue anymore. It is every sector.
The question for schools is whether students leave with the Layer 1 and Layer 2 foundations they need, or whether they spend their twenties and thousands of pounds catching up through certifications that their school education should have covered.
The "New Skills Triad"
The World Economic Forum has identified what they call the New Skills Triad for modern workforce readiness. Three capabilities that define whether someone can compete.
Carbon intelligence: understanding sustainability, carbon accounting, and environmental regulation. Virtual intelligence: excelling in remote communication, digital collaboration, and hybrid work. AI proficiency: using AI tools effectively, ethically, and securely.
That triad tells you something important. AI literacy on its own is not enough. It sits alongside environmental awareness and digital collaboration skills. All three need to be woven into how schools prepare students, not bolted on as electives or PSHE fillers.
What this means in practice
A school that only teaches AI in Computing is covering Layer 3 and 4 for a handful of students. A school that builds AI literacy across every subject is covering Layer 1 and 2 for everyone.
Layer 1 lives in Geography (interpreting climate data visualisations), in History (evaluating AI-generated source summaries), in Maths (understanding probability and prediction), in English (analysing AI-generated persuasive texts).
Layer 2 lives in PSHE (ethical implications of facial recognition), in Business Studies (how companies deploy AI for hiring decisions), in Art (authorship questions around AI-generated images), in PE (algorithmic decision-making in fitness tracking apps).
The connections exist in every subject. Teachers just need someone to show them where.
That is what AILitKit does. It takes the lesson the teacher is already planning and maps the AI literacy connections for them. Four activities. Coaching language a non-specialist can use. Support, challenge and differentiation built in. Framework alignment. 1 minute, not 30 minutes.
Layer 1 and Layer 2 for every teacher. Every subject. Every lesson.
Matthew Wemyss is the founder of AILitKit and IN&ED, and author of AI in Education: An Educator's Handbook.