Tackling the AI Readiness Illusion: From Potential to Productivity

Pearson's 2026 School & College Report found 45% of secondary students and 56% of college students already use AI as part of everyday learning, numbers that could easily read as evidence the next generation is well ahead on AI skills. Dave Treat, Pearson's Chief Technology Officer, argues that's exactly the illusion in the title: using AI tools daily is not the same as being genuinely AI-ready for work, and the gap between the two is being badly underestimated.

The teacher-side numbers explain why. Only 25% of school and college educators received any AI training in the last academic year, just 18% of secondary teachers feel confident actually teaching about AI, and only 7% of secondary teachers believe the education system currently places sufficient emphasis on digital skills at all. Meanwhile 85% of students themselves say being prepared for a digital world matters to them, a rare case of the people being taught wanting more of something the system isn't yet equipped to deliver.

The employer-side data, from a joint Pearson and AWS study, shows the same optimism gap playing out one step further down the pipeline: 78% of higher education leaders believe their institutions already meet employer expectations for AI skills, while over half of employers say their primary hiring challenge is finding graduates with the AI capability they actually need. Two groups looking at the same graduates, reaching opposite conclusions about whether the skills gap has already been closed.

Treat's sharpest distinction is between prompting and strategic application: being able to get a usable answer out of a chatbot is a genuinely different, much shallower skill than applying AI strategically at an enterprise level, understanding where it should and shouldn't be trusted, and integrating it into a real workflow rather than a one-off task. He puts a number on the cost of getting this wrong: £96 billion a year, his estimate for what inefficient career transitions and skills mismatches already cost the UK economy, a figure he expects to rise as demand for genuine digital and AI skill accelerates faster than the education system's ability to verify who actually has it.

His proposed fixes are specific rather than a generic call for "more AI education": prioritise teacher training and confidence first, since a curriculum change is only as good as the person delivering it; embed AI learning directly into how students already work rather than as a bolted-on separate module; and build digital wallet-style credentials that can verify specific AI skills in a portable, checkable way, so an employer isn't left guessing what a qualification actually certifies. The throughline across all three is that the readiness gap isn't a technology problem, it's a verification and training problem sitting one level below the technology itself.