Care is not for "snowflakes"
We talk a great deal about personalization in higher education, especially with the emergence of AI.
Less often do we talk about care, however.
Most AI-coaches “personalize” by accelerating students through content by optimizing pace, by nudging productivity, or by offering just-in-time answers. This can feel supportive. But it quietly assumes the problem is speed, not preparation; efficiency, not formation.
What if we began with a demanding question: How do we provide, at scale, appropriate educational pathways for differently situated students enabling them to negotiate their respective challenges without lowering expectations?
As the authors in an Educause Review essay point out: "Colleges and universities can leapfrog from personalized to N-of-1 precision learning by modernizing their data architecture, building dynamic learner profiles, and adopting computed curriculum."
Faculty and teachers define the outcomes, the competencies necessary and what counts as demonstrated learning. What we should personalize is the route, namely, the sequencing, scaffolding, feedback, and evidence required for students with very different starting points to reach the same academic standard with dignity intact. And, we cannot pretend that any professor or teacher can personalize instruction sufficiently. But, could they do so with the appropriate GenAI navigator?
Consider two students in the same gateway course.
One arrives academically fluent, comfortable with abstraction, academic language, and self-directed learning. The other arrives capable and motivated, but carrying gaps in prior instruction and confidence.
We need not quietly lower expectations for one while rewarding speed in the other.
Both should be held to the same destination: demonstrated competency in what the course claims to teach. What should differ is the path.
For one student, we may be inclined to compress practice, surface fewer scaffolds, and emphasize synthesis and transfer. For the other, we may need to slow the sequence, re-encode concepts in multiple representations, prompt retrieval and reflection, and require additional demonstrations before progression.
