It works like that or you think it should work like that?

If browser gets zero information, it drastically reduces UX. It even does not know to suggest permission popups without being misleading in the end.

I don't think, I wrote enough code with those APIs.

You call `navigator.usb.requestDevice({filters})`. It returns a promise and browser displays a window with device list according to the given filter. User chooses device or presses "cancel". The promise resolves with single device or rejects with error. You can't fingerprint anything without user interaction. Also there's block list which does not allow to connect to keyboard, mouse, so typical user does not even have any USB devices to choose from.