The Vibes Were Off: Teacher Intuition, the Data Obsession, and the Thing AI Will Never Have

I was vibe-aware before the kids these days started talking about vibes. Though I always feel like Steve Buscemi when I hear them say it.

What I can say, and for those of you who have either seen me present, heard me talk, or chatted with me, I definitely am self-aware of the vibe I bring everywhere. Part of my inner child thinks I'm Jeff Spicoli from the end of Fast Times at Ridgemont High..

https://www.youtube.com/watch?v=80uTRD1-F5A

My tasty waves are something I find engaging, or I hope engages others. My cool buzz....probably my mental health medication and a Dr. Pepper. And I'm fine.

Because I'm mostly vibe-self-aware, I have stopped a lesson mid-activity because the vibes were off.

I could not have told you exactly what tipped me off. It was not a single thing. It was a little bit of everything at once: a shift in the room's energy, a student who was too quiet when they are usually not, a laugh that landed slightly wrong, the particular quality of silence that means something is happening underneath the surface that the lesson plan does not account for. I stopped. I checked in. I was right. (I have learned that the kids and I use "vibe check" differently.)

I have also sent a backchannel message to a kid in the middle of class, a quiet "hey, you okay?" delivered through Google Classroom while the rest of the room kept working, because something told me to. Not a data point. Not a dashboard alert. Something I could not fully articulate but could not ignore.

Every teacher reading this knows exactly what I am talking about. That feeling. That read. That moment where your body knows something your brain has not fully processed yet.

It has a name, your teacher spidey-sense, or as researchers call it, pedagogical tact. And it is one of the most sophisticated things a human being can do.

It is also something AI will never, ever replicate.

What Teacher Intuition Actually Is

Let's start with the research because I want you to understand that what you feel in those moments is not a guess. It is not a hunch. It is expertise firing.

Researchers call it pedagogical tact, a teacher's ability to handle complex classroom situations that require immediate action. A 2019 study published in the British Educational Research Journal describes pedagogical tact as the direct enactment of teacher intuition and identifies two distinct types happening simultaneously in an expert teacher's brain. They offer the following analysis in their review of the literature exploring intuition:

The first is inferential intuition, quick pattern recognition drawn from accumulated experience. You have seen this before. Not this exact situation, but something close enough that your brain matches the pattern and generates a response before you consciously decide to respond. The second is holistic intuition, the synthesis of otherwise unconnected information into a new understanding. The kid who is too quiet, the laugh that landed wrong, the particular quality of the silence; individually, none of those things mean anything. Together, they mean something. Holistic intuition is what assembles them into a signal.

Both types are happening at once. And both require something that only comes from years of being in classrooms with real students: a library of patterns so large and so deeply encoded that you can draw on it in real time without knowing you are doing it.

Gary Klein, whose recognition-primed decision model is the gold standard in research on expert intuition, found that professionals with genuine expertise engage in intuitive decision making 80% of the time or more. Not occasionally. Not as a fallback. As the primary mode of operation. Firefighters, neonatal nurses, chess grandmasters, and teachers. Experts do not deliberate their way through every decision. They read the situation, match it to experience, and act.

A recent NIH study on teacher expertise describes what researchers call discretionary spaces, the rapid-fire decisions teachers make that are not in any lesson plan. Choosing which student to call on. Deciding how to address a behavior in the moment. Determining whether to move forward with a lesson or stop and review. These decisions require pedagogical knowledge, individualized knowledge about specific students, and what researchers call meta-awareness of how the teacher is perceived by the class, all firing simultaneously, all in real time, all in service of thirty different human beings with thirty different needs.

So, mid-lesson, when I noticed so many things feeling off, my inner Spicoli pushed one of the buttons on the Inside Out console in my brain. These reactions are not just feelings. They are forms of intelligence. A hard-won, deeply human, profoundly complex form of intelligence that took years to build and cannot be downloaded.

The System That Doesn't Trust It

There is no shortage of digital ink when it comes to my sharing of why I think the system of schooling is broken. I've written about how the progression from pre-school through high school neuters curiosity. I've argued that we've used arguments about technology being bad to mask poor instructional design and bad teaching. I've written about how systemically we are treated as feedback vending machines with no regard for our emotional health as humans who teach. Today, I fire a new shot at the system. If one more person tells me to ignore my gut and look at the data instead, I will set my ever-thinning hair on fire.

The No Child Left Behind Act arrived in 2001. Every Student Succeeds Act followed. And with them came the era of data-driven decision making in education: the belief that quantifiable, measurable, standardized data should be the primary driver of educational decisions. Not teacher knowledge. Not professional judgment. Not the read of a room. Data.

Now, I want to be fair. Data has a role. Not just because I love the meticulousness of creating spreadsheet formulas. Formative assessment data, carefully used, genuinely helps teachers understand what students know and don't know. Summative assessment data allows us to evaluate our practice and determine if we need to create alternate means for assessing our students who may not test well. I am not exclusively arguing that data is bad.

I am arguing that the overemphasis on data as the primary or exclusive source of educational truth has quietly communicated something corrosive to teachers: that what you feel in a classroom does not count unless you can put a number on it. Sorry, Beyonce, we cannot put a ring on this one.

A scathing critique published by Jason Isaacs in the Journal of Educational Thought puts it plainly. The data-driven decision-making movement overemphasizes the rational at the expense of the humanist influences on education. It imposes a reductionist evaluation and measurement framework on an irreducibly human profession. He argues that an overemphasis on data is, at best, a form of illiteracy about what is truly happening in our schools. The tools of interpretation are being borrowed from technocratic and commercial sources that were never designed to understand what actually happens between a teacher and a student in a room. They were designed to cosplay as entities that care about improving the quality of education, but they just want your money.

There is also a fundamental irony baked into the data obsession. Research shows that data interpretation is heavily influenced by teachers' existing beliefs, intuitions, and professional judgment anyway. In an article published by Vanlommel et al. in the International Journal of Educational Research, they assert that teachers do not receive data as neutral facts and respond rationally. They filter it through everything they already know, which means the intuitive expertise you have built over the years is operating on the data, whether the system acknowledges it or not.

In other words, the system spent twenty years trying to replace teacher intuition with data. What it actually did was add a layer of paperwork on top of the intuition that was already doing most of the work.

The Thing AI Will Never Have

Here's the connection to a broader theme built across the last several Friday posts.

The curiosity post argued that wondering requires safety, that the brain will not open its exploratory circuits unless it feels protected and seen. The Google/AI post argued that what technology exposes is not a technology problem but a human design problem. And this post is the final piece: the argument that what makes a teacher irreplaceable is not the content they deliver but the intelligence they bring to a room full of human beings in real time.

AI can process data faster than any teacher. It can flag attendance patterns, track assessment trends, and generate early warning indicators with a speed and consistency no human can match.

What it cannot do is walk into a room and feel that the vibes are off. It doesn't have a "Spicoli mode," no matter how advanced LLM's become.

It cannot detect the particular quality of a silence that means something is wrong underneath the surface. It cannot integrate the way a student held their pencil, the micro-expression that crossed their face when you called their name, the fact that they usually sit next to their friend, and today they didn't, and synthesize all of that into a decision to stop everything and check in.

It cannot do this because intuition requires a body. It requires a history. It requires genuine stakes; the kind that come from actually caring whether this specific student, in this specific moment, is okay. AI has none of those things. It has data. Data is not the same as knowledge. And knowledge is not the same as wisdom (I've had several D and D characters that explore the difference between those two very different things). And wisdom is not the same as being present in a room with another human being and knowing, without being able to fully say why, that something matters right now.

The assumption underlying a lot of educational technology, including AI, is that if you present teachers with better data, you will get better decisions. A 2021 systematic review found that the evidence for this assumption is largely inconclusive. The research on whether data-driven technologies actually improve teacher decision-making or reduce bias is far weaker than the confidence with which those technologies are being deployed.

What the research does consistently show is that expert teachers, using their intuition, make good decisions at a remarkable rate. Not perfect decisions. But good ones. Human ones. The kind that account for the thirty different people in the room rather than the aggregate trend line.

Trust Yourself

I want to close this the same way I try to close every hard week in the classroom.

You have built something that cannot be automated. The library of patterns you carry, every class you have ever taught, every student you have ever read correctly, every moment you stopped when the vibes were off and turned out to be right, that is not soft knowledge. That is expertise. Real, studied, neurologically complex expertise that researchers have spent decades trying to fully understand and still cannot completely explain.

The system will ask you to look at the dashboard. Look at it. Data has value, and I am not telling you to ignore it.

But do not let the dashboard tell you that what you feel in a room is not real. Do not let the metric convince you that the kid who is technically fine on paper is actually fine when your gut is telling you otherwise. Do not let the overemphasis on the rational talk you out of being human.

Here is what I want you to remember the next time you feel it: that shift in the room's energy, that student who is too quiet, that laugh that lands slightly wrong. That is not a vibe check. That is your pedagogical tact firing. That is your inferential and holistic intuition assembling signal from noise in real time. That is decades of pattern recognition doing something no dashboard, no algorithm, and no large language model will ever be able to do.

The kids think a vibe check is asking if the energy in the room is good.

You have been running vibe checks since before they had a name for it. And yours are backed by neuroscience.

You are not a data processor. You are not an algorithm with a teaching certificate. You are an expert practitioner with a hard-won intelligence that fires in real time, in service of real people, in ways that no dashboard will ever fully capture.

The vibes were off. You stopped. You were right.

That is the job. And it is why you are irreplaceable.