We often talk about threat models here in the privacy community. But I just want to illustrate how different threat models can be.

There is a universe of difference between worrying about what a district attorney can use in criminal court based on evidence rules for drug-related charges that could put somebody away for 5 to 10 years in prison.

And worrying about a totalitarian murder squad picking up your friends and family and torturing them just in case they know something. When members of your cell disappear based on suspicion no evidence required.

These threat models are massively different, doing anything that even raises suspicion gets people killed... I'm not saying it invalidates our online discussions, but when the stakes are higher what people actually use, and experiment with, and are willing to put up with changes.

And by raises suspicion, I'm not just talking about an intelligence officer, I'm talking about algorithms just saying oh this account's interesting let's target it. Just like Israel is currently doing right now, to automatically identify hamas-based targets of interest. We have no idea what's going into that algorithm, it might be hey this phone has briar installed let's bomb it