[begemotz]

I think you have a great idea here and I would keep at it. There are very few, if any, good solutions to combating the influx of AI generated work in academia. For many, I think your solution would be welcomed - even if imperfect- it gives administrations something they can point to when asked what they are doing to combat academic integrity associated with AI. I am happy to talk more if it would be useful -- I can think of a few longer term features that would potentially be useful.