Does Education Really Need an Upgrade?

The world keeps evolving. Schools stand still. And in doing so, schools fail students in developing the skills they require to adapt.

A heads-up: this post meanders, because the thinking did. I was turning over the tension between how fast AI is evolving and how firmly schools stay entrenched in archaic teaching models. So I spent a morning thinking out loud with ChatGPT-4o, trying to reconcile the gap between what we say school is for, preparing students for their future, and teaching habits that keep them stuck in our past.

Anne Collier had pointed me to a LinkedIn post by Rebecca Winthrop. I had already listened to the Ezra Klein podcast when it dropped, but revisiting it felt timely given all the recent noise around AI and education. So I popped the link into ChatGPT-4o for a quick review.

One thing led to another, and the next thing I knew I was down a rabbit hole with Marc Prensky’s 2014 white paper, The World Needs a New Curriculum.

It was fascinating to look back at Prensky’s theories while still chewing on a recent article by Mustafa Suleyman, We must build AI for people; not to be a person. Prensky was right. We do need a new curriculum.

That led me to dust off an old video from 2009. My students and I helped Peggy Sheehy and her students create a keynote for the Net Generation Education Project. The script was written entirely by Peggy’s kids. My students created the machinima (the animated clips) and I helped edit the video. It was a fun collaboration between our two schools, and a huge success as a keynote.

It has been 16 years since that video was made, and the students’ call to action is still unanswered. Maybe one anecdote sums it up. When a highly regarded education consultant asked Peggy’s students to cut their video down to one minute so he could share it with teachers around the world, they refused. Their reply: “We listen to teachers six hours a day. We think teachers can listen to us for six minutes.” Out of the mouths of babes...

Watch “No Future Left Behind” and ask yourself if it is still relevant. Analyse whether what those students asked for has been enacted. Not yet? Then if not now, when?

https://youtu.be/kra_z9vMnHo?si=leQh1ND8_-aC8rY8

Keynote for Net Generation Education Project - Sixteen years later, students are still waiting for an upgrade

All of these threads pull at the same persistent question. Why do schools consistently fail to evolve with the world? Why are we educating students for an archaic age that no longer exists? Are we being negligent not to keep up?

I engaged AI to help me think through these questions. What follows is not the conversation itself. It is what the conversation surfaced for me. I have archived the full chat as a PDF, a snapshot of where my thinking, and the tool, sat in September 2025. You do not need it to follow the thread below. It is there if you want to walk the whole route with me.

Does Education Really Need an Upgrade - AI Chat Transcript

Purpose of school

I asked ChatGPT-4o to drop the flattery and put on its devil’s advocate hat. I wanted a thinking partner, not a cheerleader. We started with Winthrop and Klein. Their argument is one I have been making for years. When AI can handle the rote tasks, content mastery stops being the point. What is left is the stuff that does not automate: thinking, creativity, knowing yourself, working with other people, adapting.

Winthrop also names something I see every day. Student disengagement is not a behaviour problem. It is a symptom. Kids slip into passenger mode when the curriculum has nothing to do with their lives.

Revisiting Prensky’s work

That sent me back to Prensky. His core move is to call school subjects proxies. Maths stands in for logical thinking, English for communication, science for inquiry. His pitch is to stop teaching the proxy and teach the underlying capability directly.

I half agree. I have spent my career refusing to treat subjects as ends in themselves. But I do not throw them out either. I do not reject proxies, I replace the broken ones with real-world equivalents. AI fluency, media literacy, ethical reasoning, these are not stand-ins for something more important. They are the literacies, full stop.

His credentialing argument lands harder for me. Portfolios and real accomplishments over transcripts and test scores. That is the self-directed learning model I built. My caveat is the one Prensky skates past: who decides what counts, and what happens to the kids who do not have the networks or resources to go find their “passion” in the first place? Ditch credentials carelessly and you simply hand the advantage back to the privileged.

Here is the part that surprised me. Twelve years ago I hated Prensky’s word “effective.” It smelled like assessment creep, the same old performance language in a new coat.

Revisiting it now, I can see he was reaching for a way to legitimise the very things that drew me to the New Zealand Curriculum's Key Competencies. The word still grates, but I can now acknowledge he was not wrong.

Schoolification

Then the conversation took an interesting turn by challenging me in a very personal way. ChatGPT-4o asked whether my own work could be schoolified.

In 2017, I co-constructed the Ako Torowhānui cline with Dan Miller, Lena Meinders and Simon McAtamney. It was created as a graphic image to provoke conversations with my students about who drives the learning and set the stage for Student-Driven Learning (SDL) in my classroom.

When I was asked to develop a school-wide SDL program with Deputy Principal Simon McAtamney, we used the Ako Torowhānui cline with teachers to provoke the same conversations. Most teachers were keen to give it a go, while a very small but vocal group of teachers consistently rejected pedagogy that took them out of the centre of directing students’ learning.

With a shift in administration, political manoeuvring gained a foothold. I found myself feeling pressure to simplify professional development to make it easier for teachers to understand the pedagogy behind SDL.

I modified the cline by adding “Noob to Boss” labelling. It was classic schoolification. The cline had been doing its job in my classroom for years, but by simplifying it for teachers, I dumbed it down for students. Adding those labels assigned benchmarks that were never intended to be the work of the Ako Torowhānui cline. I had schoolified the very core of my work by bending it back toward the system I was trying to change.

The irony is that it didn’t even work. Claims were made that the word “noob” was demeaning and disrespectful to students. It became another distraction that fed into a campaign to close the school-wide SDL program. I find that particular detail extremely telling.

I specifically chose that terminology from gaming culture. Being a noob is the doorway to learning, a threshold to cross to build skills. I always gave students the example of Richie McCaw, one of the greatest All Blacks to ever play. At some point, even he picked up a rugby ball for the very first time. The kids got it.

Alison Gopnik’s work is clear that children are wired to explore, watch, mimic, fail and try again. School is often where they are taught to attach shame to not knowing. It was the adults who flinched, not the kids.

That is the trap: “anything can be schoolified.” The colonial/industrial model is the map all of us were handed: teachers, parents, politicians, kids. When we are uncertain, we reach for the map we know. Sometimes, even me.

Is AI different?

I have watched many tools come and go. The ones that got adopted into school were all schoolified. I eventually lost faith that any new technology could disrupt, shift, or reshape our collective mindset about school, youth, tech and learning.

Still, I can’t help but wonder what schools will do with AI. They can try to ban it, provide limited access or even schoolify it. What they can’t do is keep students from accessing it outside of school. That gap is the space I'll be watching.

I have enough experience with AI to understand some of its potential, as well as some of its dangers. However, no one can truly predict how AI will evolve and reshape our world.

As a dyslexic, I know what AI already offers in customisable learning. I’m grateful to finally have a tool that can be calibrated to precisely meet my unique processing and learning needs. For the first time in my life, I feel unconstrained by the dyslexia. I cannot overstate what a game changer this is for me personally. I am excited by the potential of what it could mean for all students, if schools are brave enough to change.

Just as important, after 25 years of working with youth and emerging technologies, I'm keenly aware of the perils our students already face with AI. I feel more urgency than ever that they are given ample opportunities in school to develop their dispositions, capabilities and capacities. Honing these is the most reliable protection against unpredictable change.

Will AI be schoolified or finally disrupt the old paradigm? Let's consider that AI undercuts the whole premise of knowledge transmission that schools are built on. Let's acknowledge that students already use AI to either assist with, or offload, school assignments. Teachers do the same with their prep, marking and reporting. So is this the beginning of an evolution, or are we still simply playing school? I have no clue.

Shifting Focus

The biggest surprise in my informal research, both in my classroom and on my own, is that working with AI keeps pushing our humanity to the surface. For 25 years I framed digital tools as amplifiers, ways to make learning relevant and to create, collaborate, communicate and take action. AI is doing something else that deserves to be named.

The unexpected question that now drives my work is what it means to be human in a world increasingly shaped by digital technology. Suleyman calls it Seemingly Conscious AI, a mirror convincing enough to unsettle our sense of ourselves. With respect to human development, this is what concerns me most about AI and our youngest students. I addressed this in my post, Exploring AI: Perception vs Reality.

It is why I feel the urgency for schools to recognise we no longer have the luxury to avoid the change Peggy Sheehy’s students pleaded for back in 2009. Those students are now adults and starting their own families. We cannot fail their kids at the moment it matters most. If not now, when?

Anchors for this conversation

• Ezra Klein, NYTimes Opinion podcast with Rebecca Winthrop, “We Have to Really Rethink the Purpose of Education”

• Marc Prensky, “The World Needs a New Curriculum”

• Mustafa Suleyman, “We must build AI for people; not to be a person”

TL;DR Summary

• Sparked by Ezra Klein’s interview with Rebecca Winthrop, “We Have to Really Rethink the Purpose of Education.”

• Explores why schools resist meaningful change, despite decades of calls for reinvention.

• Uses Marc Prensky’s “new curriculum” and Mustafa Suleyman’s “seemingly conscious AI” as deeper lenses.

• Sits with the tension around student-driven learning, central to Prensky’s call and echoed in student voices from the 2009 video, yet persistently sidelined in schools.

• Surfaces how even bold ideas get schoolified, pulled back into the colonial and industrial model, and admits even I am not immune under pressure.

• Argues AI is different. It cannot be tamed into old structures and it challenges the very premise of schooling.

• Concludes that AI forces us to look beyond what we learn and how we learn it, toward what it means to be human in a world shaped by technology.

Addendum: This post was reformatted in June 2026. The full AI Chat transcript was included as part of the original post but, honestly, too much to read.

I worked with Claude (Opus 4.8) to identify the various threads of the transcript and co-write a summary of the points that mattered most.