NEW RESEARCH: The Skills AI Can't Replace

If you like this episode or it makes you think, please let us know in the comments and supporters by subscribing and leaving a review. Thank you. Today we are exploring a brand new research report from Nordanglia Education, published in partnership with Boston College, looking at Metacognition and its impact on what they call future ready skills. The report is called "Teach in the Skills AI Can't Replace" and it dropped today. I have to say this one really landed with me because it gets to the heart of something I talk about all the time. The difference between outsourcing, the doing and outsourcing, the thinking. So let me set the scene. Nordanglia is one of the biggest international school networks in the world, 89 schools, 37 countries, over 100,000 students. And two years ago, they partnered with the Lynch School of Education and Human Development at Boston College to run what they describe as one of the most comprehensive studies of Metacognition in schools to date. We're talking 29 schools, 20 countries, more than 12,000 students, 5,000 teachers, and over 500,000 individual student reflections captured through a custom platform they built specifically for this research. That is a serious data set. Now before we go any further, let's just make sure we're on the same page about Metacognition because it's one of those words that gets thrown around a lot in education without always being pinned down. The report defines it as the ability to understand how you learn best and adapt your thinking to thrive in different situations. I'd simplify it even further. Metacognition is thinking about your thinking. It's the moment a student pauses and asks themselves, "What do I already know here? What am I finding tricky? What could I try differently? It's not a subject, it's not a programme, you bolt on, it's a habit of mind." And the question this research was trying to answer is deceptively simple. What happens when you teach that habit deliberately? And this is where the findings get really interesting. According to the report, teachers observed significant gains across every future-ready skill they measured. We're talking 72% growth in collaboration, 70% in curiosity, 69% in creativity, 68% in critical thinking, 60% in commitment, 59% in compassion. Now those are teacher-reported figures, so let's hold them with appropriate respect but also appropriate scrutiny. Their perception data not test scores, but what makes them compelling is the consistency. Across 29 schools and 20 countries with thousands of teachers, the pattern held. That's not a fluke, but here's the bit that really caught my attention. The study also looked at the specific mechanism driving these results, something they call thinking routines. These are simple structured prompts that teachers embed into everyday lessons. The most well-known example they give is called "see, think, wonder". The student looks at something, a text, a problem, an image, a dataset, and asks three questions. What do I notice? What does that tell me? What questions do I still have? That's it. Three questions, nothing fancy, no special technology, no expensive platform, just a structured pause. And the data on frequency of use is striking. When thinking routines were used regularly, students showed around 21% growth in critical thinking and 20% in curiosity. But when teachers used them daily, and this is the key distinction, gains hit at least 40% across all skills, approaching 50% in curiosity in compassion. Daily, not weekly, not when you remember, daily. The dosage matters. And that tells us something really important about how these skills develop They're not taught in a single lesson. They're built through repetition, through habit, through the consistent expectation that you will pause and reflect before you act. Now let's connect this to what's happening in the wider world because the time in of this report is no accident. We are deep into the AI era. Every student with a phone has access to tools that can explain, summarize, translate, generate and suggest faster than any human teacher. The report references the world economic forums, finding that nearly 40% of today's core skills are expected to change by 2030. So the landscape students are preparing for is genuinely uncertain. And the question the report opens with is one, I think every educator and every parent is quietly asking themselves, if answers come too quickly, will my child still learn how to think for themselves? And this is where I think this research connects so powerfully to something I've been banging the drum about for years. Outsource the doing, not the thinking. AI is extraordinary at doing. It can draft, calculate, sort, summarize, translate all the doing. But thinking, that's different. Thinking is the moment of uncertainty. It's the pause. It's the decision about what to try next when you're not sure. And it's the ability to sit with a problem long enough to work your way through it rather than reaching for the shortcut. And what this Nordanglia research is showing is that metacognition is the mechanism that protects that capacity for thinking even in a world flooded with instant answers. Think about it in your own classroom for a moment. How many times have you watched a student get stuck? Not because the work was too hard, but because they didn't know where to begin. They're staring at a blank page. They're waiting for you to tell them what to do. They're looking for confirmation that their idea is the right one before they've even tried it. That's not a knowledge gap. That's a metacognitive gap. They don't lack information. They lack the habit of asking themselves what they already know and what they could try next. And the beautiful thing about this research is it shows that gap is closable. With simple, consistent, daily practices, students start to internalise those questions. They stop waiting and start deciding, you know, the student data supports this to. According to the report, 85% of students said they had a better understanding of their own strengths. 76% reported increased independence. And 72% said their knowledge of how they learned had improved. Now again, that self-report data and we should treat it as such. But those numbers tell a story about agency, about students who feel more in control of their own learning. And in an AI world, that sense of agency, the belief that you can work something out on your own might be the most important thing a school can develop. Let me push this a bit further because I think there's a non-obvious connection here that's worth exploring. When we talk about AI literacy in schools, we tend to focus on the tools, how to prompt well, how to evaluate outputs, how to use AI responsibly, and all of that matters. But what this research suggests is that there's a foundational layer underneath all of that. And it's metacognition. Because what is good prompting if not metacognition applied to a machine? You have to know what you're trying to achieve. You have to assess whether the output meets your needs. You have to decide what to adjust and try again. That's, see, think wonder applied to a chat interface. The students who are best at working with AI won't be the ones who know the most commands. They'll be the ones who've developed the strongest metacognitive habits. The report also makes a point about teachers that I think deserves highlighting. By the end of the research, up to 96% of teachers agreed that metacognitive approaches help students succeed, but not just academically, but beyond school. And the qualitative observations are telling. Teachers didn't report louder classrooms or more activity. They reported more thoughtful questions. clearer explanations, greater determination when things felt difficult. That's a different kind of engagement. It's not about excitement or energy. It's about depth. And depth is exactly what gets squeezed out when we rush to cover content or chase the next shiny tool. I want to come back to something that the report says right at the end, because I think it captures the whole argument beautifully. It says that children who will thrive in the future won't be the ones who simply know the most. They'll be the ones who can notice when something doesn't add up and pause rather than panic. Who can decide what to try next without waiting to be told? Who can stay with a problem long enough to work their way through it? Those three capabilities, noticing, deciding, persisting, are metacognition in action. And they're the skills that sit beneath confident learners and confident adults. So what's the takeaway for educators? I think it's this. We don't need to choose between teaching AI skills and teaching thinking skills. They're the same thing at the deepest level. Metacognition is the operating system. AI tools are the applications that run on top of it. If you build the operating system well, if your students can plan, monitor, evaluate and adjust their own thinking, they'll be able to work with whatever technology comes next. And that's not a prediction about the future. That's what this research is showing right now across 29 schools and 20 countries. The practical implication is beautifully simple. Start with a thinking routine, use it every day, not as a special activity, not as a Friday treat, every single lesson, see, think, wonder, or any variation that asks students to pause, reflect, and explain their thinking before they rush to an answer. The data suggests that consistency is the key variable. And the most powerful thing about it, it doesn't require any technology at all. It just requires a teacher who is willing to slow down and make space for thinking, which if you think about it, is quite a radical act in a world that's accelerating in every direction. Dr. Kate Erika, who leads education research at Nordanglia, makes a point in the press release that I think is worth ending on. She says that what stands out in this research is how meta-cognition helps level the playing field. When students understand how they learn, they gain confidence, independence, and the belief that progress is possible for them. She calls that a powerful form of educational equity. And I think she's right, because in an AI world, the gap that matters most isn't the technology gap, it's the thinking gap. And meta-cognition might be the most effective tool we have to close it. That's all for today. Thanks for listening.