The thing we need last year is regulations enforcing ethical behavior and transparency. What data were the models trained on? Did you have permission to use tha

The thing we need last year is regulations enforcing ethical behavior and transparency. What data were the models trained on? Did you have permission to use that data for that purpose? Where is your form, naturally working without third-party trackers, in which people can request their data to be removed from the training data and the learned model (hoho, that will be a challenge!)? How can people check, whether their data is in the training data? And so on.