The Guardian notes one teacher's idea of more one-to-one teaching and live lectures — though he added an obvious flaw:
"But that would mean hiring staff, or reducing student numbers." The pressures on his department are such, he says, that even lecturers have admitted using ChatGPT to dash out seminar and tutorial plans. No wonder students are at it, too.
Believe it or not there are lots of pedagogical approaches besides "assign and grade homework assignments."
And without getting in to the many alternatives, it's pretty easy to to solve this problem just by (a) all work is optional, only graded for those who want feedback (b) all evaluation is conducted by end-of-term -- or end-of-degree -- in person testing. In that case there would be zero incentive or benefit to use generative AI for assignments (other than as a study aid).
The reason that would not be exceptionally popular is because universities, rather than being forums for the great ideas of the world to be pursued by the adept, are now a place to shove listless young adults, stamp them with the correct views, and push them out with a coming of age certificate. Impediments due to lack of ability, dedication, or character are considered problems for the university to deal with, usually by finding a way to make those issues less relevant to the final outcome.
So you have to create hundreds of opportunities to get some kind of non-exam credit and half of your job is begging pleading and pushing for students to submit enough of that work to pass. And that kind of work tends to be easy to passably output with AI. But you really don't have to provide an education that way.