The Second Brain and AI: A Leadership Challenge in Education
A colleague retires. The next morning, no one knows why a project run five years ago was discontinued. The notes are gone. The context left with her.
This is not an edge case. In most educational institutions, knowledge lives in people's heads — not in systems.
What Is a "Second Brain" in Knowledge Management?
For several years, a practice has been spreading among researchers, consultants, and professionals who work with large volumes of information: the "second brain." Rather than letting notes sit in notebooks or folders that never get reopened, you connect them through a system. Each idea links to another. The whole becomes searchable, alive, held in one place.
Tools like Obsidian make this kind of structured knowledge base possible. It is not archiving. It is a living memory that grows with you.
The concept draws on the work of sociologist Niklas Luhmann and his Zettelkasten method — literally "slip-box" in German. Luhmann produced more than 70 books from a system of 90,000 interconnected notes. The core insight: what matters is not the number of notes. It is the quality of the connections between them.
How Generative AI Changes the Second Brain
Until recently, working with such a system was demanding. You had to reread your notes, recover associations you had already made, identify new ones, then build an angle. The second brain did part of the work — but your mind still did the concluding.
Generative AI changes this. You can now talk to that memory. Ask it questions. Ask it to spot contradictions, synthesize scattered readings, identify what has not yet been theorized. In minutes, it surfaces tensions between ideas that took me weeks to articulate.
That is impressive. And that is where the problem begins.
The Risk: Mistaking Speed for Depth
AI reduces the cognitive cost of synthesis. What took a day — rereading dozens of notes, pulling out the through-lines — takes a few minutes.
But processing speed is not depth of thought.
This is not a new problem. Leaders have always made decisions based on summaries prepared by others, without interrogating the assumptions behind them. AI makes the problem faster to produce — and harder to notice.
Learning research is clear on this: deep understanding requires that you build the connections yourself. When those connections are made for you, comprehension stays shallow — even when the output looks solid. For leaders, the risk is the same: delegate synthesis to AI, and you delegate part of your judgment.
Three Principles for Staying in Control
Decide what you delegate and what you keep. AI can synthesize and find connections. The decision about what matters — what deserves to be retained, deepened, held in tension — belongs to you.
Judge the quality of links, not their quantity. A useful knowledge base is not the one with the most notes. It is the one whose connections reflect your own intellectual path — not the path of a model trained on the entire web.
Protect productive friction. The best decisions often emerge from resistance, from a knot that does not come undone easily. If AI smooths that friction away, it may also erase what would have led to original thinking.
The Real Institutional Stakes
The colleague who retires takes years of context, decisions, and reasoning with her. A structured note system — shareable, transferable, searchable — can change that. Knowledge no longer rests on individuals. It becomes an institutional asset.
This is not a personal productivity tool. It is a lever for collective memory.
The LeaderTech newsletter explores these questions in depth — including concrete demonstrations of how to build a personal knowledge base, connect it to AI, and decide what to delegate and what to keep.
Source: This article is based on edition #39 of the LeaderTech newsletter on Substack (in French). Read the full version and subscribe at: https://open.substack.com/pub/christianecaneva/p/votre-deuxieme-cerveau-et-si-lia?r=rplrw&utm_campaign=post-expanded-share&utm_medium=web