Hacker News Reader

Can filters of this type be used when doing a screenshare or streaming? E.g any application that is on the screen.

Would be good to not have to worry about leaking a lot of stuff to the world.

I think there is a difficult cost benefit analysis with screen sharing. For example, what constitutes PII in that context? Maybe the text on the screen can be extracted and redacted but what about peoples faces? What about website icons in your browser tabs lists? It feels like eventually you’d need to redact everything to have privacy or accept loads of leakage

If you can run inference in real time (while doing a video call), and you can extract text through your operating system's accessibility APIs (eg the application isn't doing it's own bespoke text rendering), then probably. You'll still need to figure out where the entity appears on screen in order to censor them. (Or maybe the accessibility APIs have that information?) And you'll need some way to get in-between the OS and the screen share, like a virtual display or something.

Reply on news.ycombinator.com