Going Back to 15th-Century Pedagogy Might Be the Smartest AI Move Schools Make

There is a lot of noise right now about AI and school. Ban it. Embrace it. Detect it. Build around it. Almost every conversation treats AI as the thing that changed everything.

Here is a different read: AI didn’t change the problem. It made an old problem impossible to ignore.

The System We Inherited Was Built for a Specific Moment

Before the printing press, roughly before the mid-1400s, formal education was oral and dialogic. A mentor and a student worked through problems in conversation. The student had to explain, question, and synthesize. Not because anyone thought that was a charming philosophy. Because information was scarce, and retrieval was not the skill. Synthesis was the skill.

Then books became cheap. Schools, quite logically, reorganized around delivery and recall. Deliver the fact. Recall the fact. Test the recall. That bet made sense at the time. The student who could store and retrieve the most information genuinely had an edge.

The problem is that schools kept optimizing for recall long after the conditions that made recall valuable had changed. And now, with a phone in every pocket and a large language model that can retrieve and summarize anything in seconds, that optimization looks less like a system and more like a habit nobody thought to question.

Researchers publishing in the British Journal of Educational Technology have called for a return to Socratic, conversation-rich learning as the central structural response to generative AI. Not a workaround. Not a policy patch. A redesign of what school is actually for.

What This Means for Your Kid Today

If your child is in a typical school right now, the majority of their graded work is probably built around recall. Multiple choice is recall. Fill-in-the-blank is recall. Even most essay prompts are retrieval in disguise: find the right facts, arrange them correctly, match the rubric.

AI does all of that faster than any student ever will. Which means the question is not whether your kid uses AI. The question is whether the thing they are being asked to do is worth doing in the first place.

The skills that AI cannot replicate are the ones that were valuable before the printing press:

• Synthesis: taking information from multiple sources and building something new from it
• Judgment: deciding what matters and why, not just what is factually correct
• Dialogue: thinking in conversation, asking better questions, changing your mind when the argument earns it
• Communication: explaining your reasoning to someone who does not already agree with you

None of those skills show up on a standardized test. Most of them do not show up in a typical homework assignment. And almost none of them are taught explicitly, because the system was designed around a different goal.

The Kid Should Be in Charge of the Tool

Here is where a lot of the AI-in-school conversation goes sideways. The debate gets framed as access versus restriction. Let them use it, or don’t. Neither position addresses the real issue.

The real issue is whether the student is operating the tool or being operated by it. A kid who pastes a prompt into an AI and copies the output is not synthesizing. A kid who uses AI to test a hypothesis, argue with the output, identify what the model got wrong, and then build their own position is doing something genuinely difficult and genuinely valuable.

The difference is not the tool. It is whether the kid is in charge of the thinking.

That is not a natural state. It has to be taught, practiced, and structured. And it looks a lot more like a mentor asking hard questions in conversation than it looks like a worksheet.

Real Projects Over Passive Consumption

The practical implication is straightforward, even if it is uncomfortable for institutions built around standardized delivery.

Passive consumption, watching, reading, receiving, does not build judgment. It builds familiarity at best. The student who can sit with a real problem, work through it with a mentor or a peer, explain their reasoning out loud, and then revise when the reasoning breaks, that student is building something durable.

This is not nostalgia for the 15th century. It is a recognition that the conditions of the 15th century and the conditions of right now have something important in common: recall is not the edge. In 1420, you could not look things up. In 2025, looking things up takes four seconds. Either way, the student who only knows how to retrieve is not prepared.

The student who knows how to synthesize is.

What Globeskool Is Built For

Globeskool is an online school for kids aged 8 to 16. Core subjects, yes. But also the skills that school quietly dropped after Gutenberg: critical thinking, problem solving, communication, creativity, and working with technology in a way where the kid is directing the work, not just consuming it.

The approach is mentor-driven and project-based, because those are the conditions under which judgment actually develops. Not because it is a trend. Because the historical record, and the current moment, both point in the same direction.

If you want to see where your child stands on the skills that will matter for the next twenty years, the free assessment takes about 5 minutes.

Take the free assessment