We are not ready for the flood of AI-written fraudulent papers that will be increasingly difficult to distinguish from work that actually took place.

We need data chain-of-custody tools. Anybody using anything like that?