Yeah if it’s just a basic input with a function call for spam or not spam the risk is low. What’s the worst case outcome, it tricks it into saying no it isn’t a scam and you have to delete it manually? Hahaha

To be fair, a centralized service with your own layer of encryption on top is an option. Like Borg supports that. Not a real option for miscellaneous small file sharing or organization though, but there may be an OSS solution to that.

These statistical models are not capable of that. They are capable of self reinforcing s feedback loop of looking malicious but it’s ultimately just how the numbers align and has nothing to do with actual intent.

The benefit here is the software by design can’t read your traffic - it’s using SNI to route, not the host header, so SSL pass thru keeps the payload private. It’s billed as shareable for that reason.

Also this is open source and you can host it yourself.

Edit: I concede BakedCatboy’s point.

In either case I would probably just use FRP.

Just automated bullshit to try and detect and DMCA “infringement” automatically.

Unfortunate. Based on what I can find it’s a solo outfit, so the owner / operator (who I did find but won’t name, just check the blog though) might be indisposed for whatever reason. Couldn’t find anything online though.