A 2026 global survey puts student AI adoption at 92 percent — while only 15 percent say it shows up in their courses, and just 29 percent trust instructors to guide them on it. What campuses should do about the gap.
Ninety-two percent of students already use it. Most professors still don't know how to teach around it.
Here is a question every provost should be able to answer by now: what share of your students used AI to write, study, or solve a problem set last week? Most can't answer. The honest ones will guess low.
Picture a professor during office hours, reading a strong essay from a student who has never written like this before. The paragraphs are clean. The argument holds together a little too well. She isn't angry. She's just trying to figure out what she's actually grading anymore, and whether her syllabus still describes the class she's teaching.
That moment is playing out in offices across every campus, because the shift already happened. It didn't wait for a task force.
A global survey from the Digital Education Council covered more than 45,000 students and faculty across 35 countries. It put a number on what professors were already sensing: adoption is close to universal. Other 2026 research on AI in higher education puts student use as high as 92 percent. Course design has not caught up. In the same survey, only 15 percent of students said AI shows up in many of their courses. Forty-three percent said it shows up in none.
That's the widest gap in the data, and it's the one worth sitting with. Students are not waiting for permission. Institutions are still writing the policy memo.
Students proved they'll adopt a new tool without being asked, across every major and every campus. Institutions have not proven the same thing yet, and the gap between the two is where most of the risk sits.
The deeper problem is trust. Only 29 percent of students believe their instructors are well equipped to guide them on using AI well. In the US and Canada, that number drops to 17 percent. Students have the tool. Almost nobody has told them, credibly, how to use it.
I've spent the past two years inside this exact gap. I advise the California State University system on AI as an Innovator-in-Residence, and the pattern is consistent. The campuses that move fastest aren't the ones with the strictest ban. They treat this as a teaching problem before it becomes a policy problem.
This isn't the first time a tool arrived faster than the curriculum built for it. When calculators showed up in math classrooms, schools spent years arguing about whether students would forget how to do arithmetic. They didn't ban the calculator forever. They rebuilt the test. Assessment shifted from "can you compute this by hand" to "do you understand what the computation means," and that redesign, not the ban, is what actually protected the learning.
AI is forcing the same redesign, on a faster clock. A syllabus written two years ago already describes a workplace that doesn't exist anymore. That's the curriculum half-life problem: the useful life of a course description is shrinking every semester, and most institutions are still updating on the old schedule.
This is not only a campus story. Every household with a kid doing homework tonight is asking the same question as a dean. Who is teaching that kid to use it well? The boardroom and the kitchen table are running the same experiment, at different scales. Neither one gets to opt out.
None of this means lowering the bar. It means moving the bar to where the value actually is. Assignments that reveal how a student thinks hold up no matter what wrote the sentence. Faculty who use these tools openly earn more trust than faculty who pretend the tools don't exist. Both changes cost real time. Neither one requires a task force to start next semester instead of next year.
Some colleges will treat this moment as a chance to teach judgment, not just compliance. Their graduates will know how to direct these tools at work. The colleges that treat it as a discipline problem will spend the next three years writing policy for a fight that's already over.
If you lead a department, a campus, or just a household with a kid doing homework tonight, the question isn't whether AI belongs in the room. It's who's teaching them to use it well. I'm easy to find if you want to work through what that looks like on your campus.
Sources: Digital Education Council — AI in Higher Education Global Survey 2026; AI in Higher Education LATAM Survey 2026.
Alex works with executive teams at global enterprises on AI strategy, governance frameworks, and organizational readiness. Available for keynotes, C-suite workshops, and advisory engagements.
.svg)
