What a load of FUD, sort of.
So, how does this virus get itself run in the first place?
My AV reports on files containing viruses before I even open them, blocks email with viruses as the mail client downloads them and I believe it scans incoming http type traffic for me as well (although I haven't tried to check that).
So, bearing in mind that the AV software still has most of the attack vectors covered, the fact that a program that hasn't already been detected as a virus isn't detected as a virus when it runs is hardly surprising, is it?
For those of you concerned about how easy it is to "fool" your AV software, even "clever" AV software, look for the EICAR test virus. Download it to your PC and try changing a few bytes of the "text message" portion within it.
Last time I tried that, about 2/3rds of the AV software I tried it with were completely fooled.
The few that weren't fooled did manage to be confused by simple tricks like renaming the file from .com to .txt. If you wanted to, you could probably put the file in an NT "stream" and hide it quite well.
Basically, you're screwed. It's only the incompetence of virus writers that has kept everyone safe so far but with the financial gain element of viruses becoming easier to realise, I think it's going to get a LOT worse before it gets better.
I don't trust the AV writers to be the solution. It's down to the OS and the privilege models that they use.
That's why Linux is generally immune, not because it's in some way better, but just because you don't need to run everything as root. The difference between user and system are very clearly defined. So, you can screw up one user account, big deal. Just delete it and move on.
The problem now with Linux is that there are more and more non-technical users who don't understand the model and still insist on installing packages with sudo instead of putting them in their own home dir. Encouraged by the infrastructure of apt/yum/etc....