Similarly, a gun would need to have a failure rate somewhere in the order of 1% of the typical misfire rate. E.g., if a particular gun & ammunition pairing misfires 1 time out of 1,000 then the security system must fail no more than 1 in 100,000 times or thereabouts.
At that point, any misfire due to the security system would be indistinguishable from a misfire that would have happened anyway.
A 1% increase in misfires would not equate to a 1% increase in unintentional deaths, but a viable safety system would significantly decrease them. In fact, it might be possible to mathematically derive a much higher break-even point, but I daresay that would not work out well politically.
I carry a 1911, which is a design more than a century old. I've fired that gun around 15,000 times without a single failure of any kind.
John Moses Browning was the Leonardo da Vinci of gun design. In addition to the M1911, several of his other designs are still in production, and indeed, in active duty use by various militaries.
I carry my SW1911 90% of the time these days. I occasionally switch out for a Hi-Power (another Browning design!) or a Glock 31.
The Hi-Power is honestly a more practical gun, but the trigger is harder to make decent, since Browning was working around his own patent that he'd sold. My 1911's trigger is the best I've ever seen, bar none.