The documentation states that only the lo-order bit is reliable : https://docs.microsoft.com/en-us/windows/desktop/api/winuser/nf-winuser-getasynckeystate

but https://cs.chromium.org/chromium/src/third_party/webrtc/examples/peerconnection/client/main_wnd.cc?q=GetAsyncKeyState&sq=package:chromium&dr=C

uses 0x8000 as a mask (extracting the hi-order bit).
 

The documentation states that only the hi-order bit is reliable : https://docs.microsoft.com/en-us/windows/desktop/api/winuser/nf-winuser-getasynckeystate

but https://cs.chromium.org/chromium/src/third_party/webrtc/modules/audio_device/win/audio_device_core_win.cc?q=GetAsyncKeyState&sq=package:chromium&dr=C

uses 0x1 as a mask (extracting the lo-order bit).