There are, the authors estimate, 150 Russian remote nuclear launch sites and 70 in China, approximately 2,500km (1,550 miles) from the nearest border, all of which could be reached by US air-launched JASSM and Tomahawk cruise missiles in a little more than two hours in an initial attack designed to prevent nuclear weapons being launched.
Emphasis mine, I’m pretty sure even Russia can notice hundreds of cruise missiles are heading directly at their silos and figure out that this looks like an attack on their strategic nuclear arsenal in two whole hours, given that ICBMs take around a quarter of that from launch to impact.
And even if these cruise missiles were completely undetectable, it would still fail unless the strike takes out 100% of the enemy nukes. If even one is able to survive, you risk a nuclear holocaust.
Being able to theoretically wipe our all the enemy nukes without using any of your own is strategically nice to have, but on its own it isn’t enough to negate the threat of a nuclear exchange. At best, it should make your enemy more reluctant to retaliate with a nuclear launch, assuming they realize that they aren’t getting nuked and that a launch would potentially change that.
The whole idea of nuclear ICBM warfare and MAD is that you are prepared to launch at least some of your missiles before your attack capability can be removed. So, to maintain MAD capability, at least some of your missiles have to be launch capable at any time in order to effectively respond to a first strike. Of course, that readiness level can be increased if the perceived threat is higher. What that means is that a response strike needs to be able to launch in less than 30 minutes. Two hours is very generous. The first strike advantage is that you can launch most of your missiles. The MAD doctrine assures that all victories on this stage are pyrrhic.
