As I understand it, the problem is one of hyperbolic doubt. Imagine if you were on trial where the evidence against you came from, say, a computerised breathal

As I understand it, the problem is one of hyperbolic doubt. Imagine if you were on trial where the evidence against you came from, say, a computerised breathalyser. You might argue that the code has bugs. Then that the configuration was wrong. Then that the hardware design was faulty. Then that the individual components were broken. Then that the data was altered in transmission. Then that a hacker got in. Then that a cosmic ray corrupted the data. And so on. The defendant has no proof that the evidence is faulty, and the producers of the evidence can demonstrate (to some degree) that their hardware & software are correct. At some point, the court has to accept the evidence presented to it. In the UK, at the moment, the assumption is that a computer is operating correctly unless shown otherwise. The BCS (I'm a member) goes in to a little detail about why they think this presumption should change https://www.bcs.org/articles-opinion-and-research/computer-is-always-right-law-must-be-urgently-reviewed-to-stop-another-post-office-scandal-says-professional-body/ https://www.bcs.org/articles-opinion-and-research/computer-i...