White worms have been discussed at great length, I've been a part of some of those heated discussions.  What it boils down to is access.  By creating a "white worm", you are accessing and changing a system you do not have permission to do so.  The debate started long ago with Code Red (http://en.wikipedia.org/wiki/Code_Red_%28computer_worm%29) and the theoretical Code Green worm.

The biggest one that sticks out in my mind is Welchia (aka Nachi http://en.wikipedia.org/wiki/Welchia), which exploited Microsoft security flaws in order to patch them.  In turn forcing reboots.

Other worms have similar concepts, but mainly used to closing the backdoor once they get in, like Conficker and Slammer.  There are also some macro viruses which have really advanced this idea for EVIL.

Thanks to my co-worker for collaboration on this.

I had the same thoughts, adamj. Whatever your intends might be and how good they are, there are way too many possibilities and combinations where you can't tell if everything would really go as planned. The exactly behavior in each situation is not possible to foresee and therefore one might led to some sort of crash or disaster. If such a scenario would happen, even if it was very rare, would the authors (government?) take responsibility then?