No matter what one does, GPS can always be jammed eventually.
But one can make it very hard to do.
Rooftop aerials (essential for a decent reliable signal, anyway) help a lot from ground-based sources and also with interfering sources on the same metal aircraft (e.g. DME aerials being on the bottom). Anybody can do this, and everybody should. If I was the CAA, and I wanted to formalise GPS somehow, I would make rooftop aerials mandatory. It's relatively cheap. Anyone using a handheld with an integral aerial, inside an all-metal cockpit, is asking for trouble. Unfortunately these units do work 99% of the time
Introducing IRS data into the GPS receiver enables one to make it far more resistant to jamming - even if the IRS gyro has poor long term accuracy. Reasonably priced FOG-gyro attitude instruments already exist so this isn't far away.
Aerial polarisation is another technology not yet available to civilian users.
Other anti-jam technologies require more processing power but you bet the military have them. It's pretty easy to work out how to do that, too.
One essential point is that if a modern GPS is unable to compute a good solution, it says so. Identing a VOR or NDB or DME just tells you you've got the right frequency tuned; it says nothing about the instrument being useful. I've flown with VOR and DME kit which could be made to read anything, and yes they were within the DOC.
But I bet ILS will be with us for decades to come
What I don't get is why so many people keep assuming that GPS usage implies using it as a sole nav reference....