The trouble with percentages like this is that the last <1% are difficult to achieve but provide the most value. If you're a teacher looking to prevent cheating by students, you can't take the chance of falsely accusing one student out of 100.
It doesn't matter if a very particular model $A$ under particular conditions can be identified with another particular model $C$. That can very easily be circumvented by using some other model (Vicuna, GPT-4, Claude, etc) in different conditions to fool the model $C$. It's a constant race that will never be won and shouldn't even really be attempted.
One idea for determining if the content is generated by OpenAI that requires far less work is to for OpenAI to just make a database table of (document_id_hash, text_shingle_hash, shingle_weight_by_frequency) to make a system like old plagiarism detection. It's incredibly fast and simple, so uses far less compute than a big ML model to do an approximate but worse method.
Yes, yes, the API, other models, etc. But that was already a problem (see above).
Further, i'm curious how universal the accuracy is...do certain subsets incur more of the burden?
I understand you're not saying it's impossible, but is it really that much of a pain for a real human to just read the paper?