My oldest daughter has a quality that took me a while to fully appreciate. She is relentless about her goals, straight A's, serious about tennis, always working. You'd expect someone like that to be confident to the point of stubbornness. And she has that confidence. But she also has something that is rarer and, I'd argue, more valuable: she knows what she doesn't know.
When a coach gives her a correction, she doesn't bristle. She listens. When a teacher marks something wrong, she doesn't look for reasons the teacher made a mistake. She goes back to the work. There's no performance of humility there, no false modesty or social politeness. It's something more operational than that. She genuinely treats the gap between where she is and where she wants to be as useful information rather than a threat to her identity.
I didn't teach her this. I've been trying to understand where it comes from. The samurai had a word for it. They called it Rei.
In most Western translations, Rei gets rendered as “respect” or “courtesy.” Both are correct but neither captures the full weight of it. In the Bushido tradition, Rei was less about social politeness and more about an orientation of mind: the willingness to honor what is truly honorable in others, to show reverence for wisdom wherever it appears, and to hold yourself with enough humility to keep learning from everyone and everything around you.
Nitobe Inazo described Rei as the outward expression of an inner state, not manners as performance, but manners as philosophy. The samurai who bowed deeply did so not to appear respectful but because they genuinely understood that the person before them might know something they didn't. The bow was the posture of a learner.
There's a related concept in Japanese culture called shoshin, beginner's mind, that lives inside Rei. The idea is that the expert who approaches their craft with a beginner's openness sees more possibilities than the expert who approaches it with accumulated certainty. The greatest swordsmen didn't practice as if they had mastered the sword. They practiced as if the sword still had things to teach them.
This is the part of Rei that gets lost when we translate it as mere courtesy. It isn't about being polite. It is about being genuinely, persistently open, to correction, to new information, to the possibility that what you believe right now might be incomplete or wrong. That openness, practiced daily, is what allows a person to keep compounding in a world that keeps changing.
AI systems are extraordinarily confident. They deliver answers in clean, authoritative prose, with no hedging, no visible uncertainty, no acknowledgment that they might be wrong. The experience of using them is the experience of consulting something that always sounds like it knows. It doesn't flag its gaps. It doesn't model intellectual humility. It performs certainty as a default, even when the underlying data are thin or the question has no clean answer.
This matters for children who are growing up using these tools as their primary reference point. The model they are absorbing, consciously or not, is that knowledge sounds like confidence, that uncertainty is a malfunction rather than an honest epistemological state.
A 2026 review published in Child Development Perspectives examined intellectual humility in adolescence directly and found that today's teenagers face particular obstacles to developing it, including social pressures to appear confident, susceptibility to misinformation, and cultural environments that increasingly reward certainty over curiosity.[1] The paper noted that intellectually humble adolescents were more likely to question dubious information, less likely to endorse conspiracy theories, more open to opposing perspectives, and more socially connected. In short, the kids who stayed genuinely open did better on almost every measure that matters.
A separate body of research shows why this quality is so powerful for learning specifically. A study across five experiments found that high school students with higher intellectual humility consistently showed stronger mastery behaviors: they sought out challenges, persisted longer after setbacks, invested more effort to learn things they had initially failed at, and were more willing to engage when confused.[2] Intellectual humility is not the opposite of ambition. It is one of its key engines.
The most direct lever the research identifies for building intellectual humility in children is the most available one to every parent: modeling it yourself.
A 2024 study across five experiments with hundreds of adolescents found that when adults in authority openly admitted confusion, acknowledged mistakes, and expressed uncertainty, students became more engaged, more willing to ask questions, more interested in learning, and more likely to improve their grades over time.[3] The mechanism wasn't inspiration. It was permission. When the adult in the room demonstrated that not knowing something is a normal part of learning rather than a failure, students stopped treating their own ignorance as something to hide.
One afternoon I was driving my two younger daughters to their tennis lesson. We stopped at an intersection. On the corner sidewalk I could see an elderly woman sitting with two kids, maybe 8 or 9 years old. Something wasn't right about the scene. The thought crossed my mind to roll down the window and ask if they were okay, but we were already late for the lesson. When the light turned green, I drove.
Then immediately I knew I'd made the wrong call. Not just about the right thing to do, about what I was showing my kids. So I made a U-turn and went back. It turned out the grandmother had fallen. The grandkids had already called their parents and help was on the way. Everything was fine. But I told my daughters directly: I was wrong. I should have acted when we were sitting at that light. The fact that it worked out doesn't change the decision I made. That's on me.
The second strategy is creating deliberate space for “I don't know” in your household. When your child asks you a question you can't answer, say “I genuinely don't know” and then model what you do with that gap, look it up together, find someone who knows, sit with the uncertainty for a moment before resolving it. The process of handling not-knowing is what you're actually teaching.
The third strategy is protecting your child's curiosity from the answer-delivery model that AI makes so easy. There is a difference between using AI to find an answer and using AI to explore a question. One trains the habit of getting information. The other trains the habit of thinking. The parent who occasionally asks “What do you think the answer might be before we look it up?”, and means it, and actually waits, is doing something that compounds over years into a fundamentally different kind of learner.
This practice works for kids of any age and takes about five minutes. Once a week at dinner, each person at the table names one thing they were wrong about that week. Not a mistake in the sense of doing something bad. A belief they held that turned out to be incorrect, an assumption that didn't hold, something they thought they knew and discovered they didn't. The rule: no embarrassment allowed. Being wrong is the premise, not the punchline.
The point of the practice is not to catalog errors. It is to build the habit of noticing when your beliefs update, and treating that update as something worth naming. Children who practice this regularly develop a different relationship with being wrong. They stop treating it as something to hide and start treating it as something to report.
My youngest is extraordinarily good with language. Quick, sharp, naturally witty. The flip side is that she will sometimes say something she thinks is funny or clever that lands harder than she intended, a comment to a sister that crosses the line from teasing into something that stings. This happened more than once. I tried to explain that what feels witty to the person saying it doesn't always feel that way to the person hearing it.
Recently I've noticed something shift. She'll say something, and before I have a chance to say anything, she'll read the change in her sister's expression and stop herself. She'll apologize. Not every apology gets accepted immediately, her sisters are human too. But she's doing the work. She noticed she was wrong. She said so. She's trying. That's Rei in a seven-year-old. It doesn't look like a bow. It looks like a kid watching someone's face and deciding that being right is less important than being kind.
The Wrong Thing practice is designed to make that noticing a habit rather than a crisis. Start it this week. Do your wrong thing first.
The samurai who embodied Rei fully was considered among the most formidable. Not because they were more deferential, but because they never stopped growing. The warrior who knew everything had closed themselves off to the teaching that every encounter offered. The warrior who stayed genuinely open compounded. Every conversation, every correction, every challenge was material.
AI is the most confidently wrong thing ever invented at scale. It will tell you things that aren't true with the same tone it tells you things that are. It cannot model uncertainty because it was not designed to. It cannot say “I might be wrong about this” in a way that carries any real weight.
Your child can do something AI cannot. They can genuinely sit with not knowing. They can feel the discomfort of being wrong and choose to stay open anyway. That quality, built through ten thousand small moments of intellectual humility, becomes one of the most valuable things a person can carry into a world where the answers are everywhere but the wisdom to evaluate them is increasingly rare.
The challenge for this week: try the Wrong Thing at dinner. Start it yourself. Say your wrong thing first. Watch what happens.