I would place the burden of proof on showing that the firearm is safe.
That isn’t really possible with any firearm. You can certainly test some firearms in a variety of tests. Assuming no failures, then you would know two things:
1. For the tests you performed, there were no failures. You would still have no idea if there were other tests in which the firearm might fail.
2. You could say that for the tests done, there would be a specific probability that all of the firearms of that brand/model would be safe. That probability would not be 100%.
History is littered with examples of human engineering that were deemed “safe” that later turned out not to be…
Side note: I do not own or plan to own any Sig 320s. I do own other Sigs, and I’m not interested in the 320 simply because there are a lot of other pistols are more interesting to me.