Many in the firearms world feel that you should be able to shoot 200 rounds of your self-defense ammunition without any problems prior to carrying the firearm. That is a failure rate of <0.5% So for reliability a failure rate of <0.5% would be required.
I think you are misconstruing that recommendation. I have always taken that as a preliminary number only; shoot 200 to catch any obvious manufacturing defects and then put the gun on probation, and shoot a whole lot more practice ammo through it to catch any lower-rate failures. It does not mean that a MRBF of only 200 rounds, or a failure rate as staggeringly high as 0.5%, would be anywhere close to acceptable; it's just a recognition of the fact that 200 rounds of carry ammo can start to approach the cost of the gun itself. If your car's brakes could be expected to completely fail in a panic stop with a 1/200 probability, you'd get that car serviced stat, and probably wouldn't consider it safe to drive until that problem was fixed.
In my gun safe, a gun with a demonstrated overall system reliability of at least .9995, with any failure modes immediately addressable by a tap-rack-bang, is on probation; if it has a gun-related failure even once in 2000 times with quality ammunition, I'll sell it or consign it to range-toy-only status, and if it was an obviously ammo-related failure (e.g. no priming in the primer cup, out-of-spec case), I'd consign that ammo to practice only. Unfortunately, failure rates for electronically-locked guns are vastly higher than that currently, and the failure modes are often catastrophic (i.e. total loss of system functionality). Also consider that the electronics reliability has to be higher than .9995 in order not to bring the overall system reliability down to .9995.
And yet they are one of the most trusted sighting systems despite possessing all the potential flaws of a smart weapon. It simply hasn't been an issue, which is my point. But we still have folks who claim electronics do not belong in a gun under any circumstance because they will fail --.gov killswitch or no.
There are a couple of key reasons why Aimpoints are trusted more than internal electronic locks. First, failure mode; if an Aimpoint fails, the rifle is still instantly and absolutely usable; one of the selling points of the Aimpoint is that you can cowitness your backup sights through it, and just lower your head a little to shoot as if you weren't using an Aimpoint. Second, Aimpoints have a demonstrated a MRBF in the millions of rounds, not the low hundreds.
Look at the troubles Eotech is currently having, even with an MRBF of tends of thousands and benign failure modes, or the probability-of-failure thresholds used in automotive recall or aviation-safety reliability criteria. People simply do not tolerate much likelihood of failure in safety-critical devices.
I also forgot to add the absolutely most critical of all criteria for electronic handgun design;
Must not rely on continual updates, contracts, upgrades, or replacement for continued function at a rate significantly (predictably) higher than the mechanical parts. No "planned-obsolescence" will be tolerated. Google/Apple/Msoft need not apply. Upgrades to add function are perfectly acceptable and expected, but if needed for basic function, constitute a severe liability.
I'd add that the system
must not be passively or remotely updatable. If OTA updates can be pushed to the gun under any circumstances, then forget it. And it must not be capable of remote disabling by any means, legal or illegal, that would not disable a conventional firearm. I also expect it to last for decades, or be user-convertible to a non-locked firearm if not.
Someone who has conventional firearms for defensive use, and has a "smart gun" only for the "cool range toy" factor, may not have the same criteria. But most of them seem to be aimed at the defensive market, and that is a much higher bar.
gun owners are safety-conscious often to the point of obsession --they will snap this up like candy (same as how S&W Lemon Squeezer revolvers were snapped up by being designed to be slightly more difficult for a child to cock/fire)
Reliability is equally important (or more, since you can create superior safety by exterior means, e.g. a safe and a holster. The internal electronic internal lock mostly provides incremental benefit to those not storing guns in safes or carrying them on the person, at the (current) cost of vastly decreased system reliability.