The Room The Adults Are Late To

Earlier this month, the Digital Education Council published its 2026 AI in Higher Education Global Survey, one of the largest datasets ever assembled on the subject. Forty-five thousand responses. Twenty-seven thousand from students, eighteen thousand from faculty, across thirty-five countries.¹

The headline numbers were not the surprising part. Eighty-eight percent of students now use AI in their learning. Seventy-seven percent of faculty use it in their teaching, up sixteen points from a year ago.¹ Any teacher who has walked past a laptop in a lecture hall this spring already knew as much.

The number that landed harder sat quietly a page inside. Only twenty-nine percent of students believe their instructors are equipped to guide them through the technology.² Fifty-seven percent report that their assessments come with inadequate AI guidance. Thirty-one percent of faculty agree they were meaningfully involved in shaping AI policy at their institution.² The report's authors offered a sentence that reads clean even months from now. AI has moved into the mainstream of student and faculty life faster than institutions have been able to respond to it. Adoption is now widespread. Coherent practice is not.³

What The Numbers Are Actually Telling Us

The story is not that students are cheating and faculty are behind. The story is that the room has shifted underneath the whole arrangement. AI arrived in the pockets and browsers of every student in the world faster than any pedagogy or policy knew how to catch up. What the survey names is the shape of that delay.

Every previous wave of classroom technology, calculators, internet search, Wikipedia, Grammarly, arrived slowly enough that the arrangement between a student and a teacher could absorb it and slowly re-define what counted as thinking. There was time to build the vocabulary.

This one is faster. The gap between a student pasting a prompt at eleven at night and a faculty senate ratifying a policy about it is now measured in years, not semesters. The professors in the DEC survey are not lazy. They are running a course they designed when the rules were different, and the rules have changed twice since the syllabus went to print.

What Coherence Would Look Like

The word the report chose is worth staying with. Coherence. Not compliance. Not enforcement. Not a detector. Coherence is a smaller and more human thing. It is a teacher and a student in the same room agreeing, roughly, on what the work is for and what parts of it belong to whom. It is the professor who can say on Monday morning that this assignment is meant to help you think, and that the tool you might use to skip the thinking is the wrong one for exactly this reason. It is the student who can hear that because she trusts he already knows what the tool can do.

That kind of coherence does not arrive in a policy document. It arrives in tens of thousands of small conversations, one at a time, held in offices and after class, where the professor turns out to have looked at what the tool actually produced and to have thought about what it costs.

The DEC survey suggests those conversations, at the moment, are not happening. Not because faculty do not care. Because they do not yet have a way to see, in enough detail, what is happening on the other side of the assignment when the student is at her desk with the tab open.

Where The Record Comes In

This is the part of the picture we spend our days on at Koan. Not because we think we can build coherence for a professor. Coherence is his to build. But because the conversation the survey is asking for depends on something specific. A record of the student's actual work. Not the final answer she pastes into the course site at eleven forty-seven at night, but the eight minutes before that, when she wrote a sentence, deleted it, asked the tool a question, took part of the answer, and started again.

Right now, in almost every institution the DEC surveyed, that record does not exist. The professor sees the paper. He does not see the process that produced it. When the survey says only twenty-nine percent of students think their teacher is equipped to guide them, part of what it is saying is that the professor is being asked to guide something he cannot see.

A visible process changes that conversation without changing the tools. The student keeps her AI. The professor keeps his classroom. What arrives in the middle is a shared record that both of them can look at and talk about. The coherence the survey named does not require a moratorium. It requires a light switch.

The Twenty-Nine Percent

Higher education will spend the fall arguing about policies. K through twelve, still working through Ohio's July mandate and New York's summer purchasing freeze, will spend it arguing about the same. The DEC number, the twenty-nine, will show up on every side. It is a low number, and it is the correct one to work from.

Every institution reading the survey has the same choice. Push the number up by hiring more professional development, writing more policy, buying more detectors. Or push it up by making the work visible enough that the guidance becomes possible again. The first is faster. The second is what the survey, quietly, is asking for.

If the students are already in the room, and the adults are late, what do the adults bring with them when they arrive?