The media are in a feeding frenzy over this. However, the malware in question is several years old. This started I think with the FBI blowing their own trumpet over taking over a botnet (that had been working for several years).
Then the UK's National Crime Agency decided to use it as a public awareness campaign to try and get people to improve the safety and security of their PC's.
Then the press got it and did their usual thing of totally twisting it out of recognition.
Indeed, the later variants of this malware are, at least to a degree, not even dependent on their command and control servers. They are capable of talking peer-to-peer.
So the risk is inflated by the press. However, that isn't to say that this type of malware doesn't present a real and present danger to anyone using the Internet - it does.
So the usual mitigations should be applied:
- Good anti-malware software with rapid updates (the Microsoft AV seems to have fallen off the edge of acceptable recently) - Avast, etc.
- Limits on running executables - this may well be the best protection all round. Unfortunately, it isn't easy to achieve outside the enterprise & even in the enterprise, it does annoy people! Only recognised and authorised executables should be able to run.
- Outside the enterprise, sensible precautions are running Microsoft EMET and running as a standard user not a user with admin rights.
An occasional sweep with alternative anti-malware tools and rootkit detection tools may also be sensible though many of the vendors of what were just AV tools now include additional checks as well.
Oh, and inside an enterprise, there should also be additional checks at network boundaries by using threat management tools which can detect suspicious activity both coming into and going out of the network and blocking & reporting it.
Finally, the malware in question is most certainly detectable by the right tools. A quick search will reveal that.