Yes that was my argument: your approach is simply not realistic. Security people always say "just fire everyone who doesn't comply", but in the real world they rarely have that sort of power, nor should they.
If the purpose of your company is to guard something, then fine. That's unlikely to be a Silly Valley startup. If the purpose of your company is to invent something cool, something worth guarding, then the inventers are more important than the security guards. Thus you want to work with human nature, not against it.
Heck, even in the military where you can go beyond firing people who don't follow the rules, it's still common practice to accommodate human nature to improve security. The more realistically you model the human parts of your solution, the better you'll do.
(You might notice that for example, the door to the smoking area already have the best security in the world: a tight group of people who all know each other, will recongnize anyone who doesn't belong, and who pay attention to who comes and goes. But your goal seems to be "control" not "security" - the fewer people like that with jobs, the better.)