A programme team accepts a change to their assessment brief from someone who has never taught the course, on the strength of a five-minute conversation. A year earlier, someone else in the same room proposed almost the same change and got nowhere. Same idea. Different person carrying it. That is what decided the outcome.
We don't reason our way into new behaviour. We act, uncertainly and messily, and then make sense of it by telling a story about what happened. I keep coming back to this when I watch institutions respond to AI and other challenges. Sensemaking is retrospective. Strategy is prospective. The gap between them is where people get lost.
The language of "top-down" and "bottom-up" runs through almost every conversation about change in higher education. It carries implicit moral weight: bottom-up sounds authentic and practitioner-led, yet also ungoverned and hard to scale; top-down sounds imposed and procedural, yet also coherent and capable of giving scattered efforts a shared direction. The distinction feels self-evidently useful.
At some point in the last two years, most universities will have convened a working group on generative AI. Some will have produced policy frameworks. Many will have run staff development sessions. A good number will have updated their academic integrity guidelines, published guidance for students, or commissioned an internal review. All of this activity is genuine, and some of it is genuinely useful.
Most universities offer their academic staff some form of teaching development. A workshop on active learning. An induction session for new starters. A seminar series that appears in the calendar each year. The intentions behind these programmes are genuine. The evidence that they change teaching practice is, on the whole, thin.
Conversations about educational technology often orbit around efficiency, but the rapid rise of generative AI has forced universities into a long-overdue reckoning with a much deeper question: what exactly are we doing when we provide feedback? If feedback is merely the transfer of corrective information, then large language models have already won. They can parse essays, spot logical flaws, and debug code with astonishing speed. However, reducing feedback to a glorified diagnostic tool misses the fundamental reality of how university students actually learn.
Three years in. That's where we are now with generative AI in higher education. ChatGPT's arrival in late 2022 feels like both yesterday and a lifetime ago. The initial panic ("How do we AI-proof assessment?") has given way to something more interesting, more nuanced, and dare I say it, more hopeful.
Step into the lobby of almost any university, and you will likely find a mission statement etched onto the glass façade. It usually speaks of "excellence", "innovation", and "global citizenship". Yet, a mere few hundred metres away in a lecture hall, the reality often feels worlds apart from those lofty aspirations.