> It will work because honest people aren't trying to get around it. I think this makes the counterargument even stronger. Let's take for granted that this is

> It will work because honest people aren't trying to get around it. I think this makes the counterargument even stronger. Let's take for granted that this isn't intended to stop a determined leaker and is just meant to prevent honest, unintentional mistreatment of sensitive data. The question is whether the false positives outweigh the true positives. This feature will impede people from getting things done in subtle but annoying ways (making it more difficult to take notes, hurting accessibility, etc). It's likely that when this is widely deployed, many big orgs are going to overuse it and enable it as a matter of course to prevent liability. Those scenarios where honest people are blocked from doing honest things for which there's no harm are the false positives in this scenario -- there was no need to prevent those scenarios, but they were prevented anyway. Now consider the true positives: we've agreed that intentional malice is not covered by this feature, and so the true positives are limited just to the smaller subset of scenarios in which honest people unintentionally mistreat sensitive data, and don't include any scenarios where data is being intentionally leaked. I suspect the number of scenarios that fall into the false positive category will be much greater than the number of scenarios that fall into the true positive category, especially so after intentional malice is excluded. So is this really a net win for anyone?