@alwayscurious That is not what I am saying at all. You need to understand your usage model and know just exactly what portions of the kernel you are using and design your update schedule around that.

And be prepared to update/reboot at any point in time, after properly testing updates.

Companies using Linux in their "uptime-critical" products usually already know all of this and can handle it just fine. If not, then they are designed and supported wrong and that's a company problem, not a Linux problem.