There's likely enough undisclosed OS kernel vulnerabilities known to state actors that this kind of attack would not be very necessary.
if you can get to the same amount of tampering risk for cheaper, you should switch to that process.
we already spend billions on fraud detection, prevention, etc each year. and probably a significant portion goes undetected too.
if you can get to the same amount of tampering risk for cheaper, you should switch to that process.
and you obviously can, you can do - let's say - a multi-factor voting, so you use multiple devices to record your intent, and you use different kinds of hardware to assess the votes.