In the middle of 2004 we saw the first real cell phone virus, named Cabir. It was newsworthy because it was the first, and since then there have been more. I dont think of myself as an expert on them, but I dont feel very threatened by them.
In mid-March Cabir (pronounced "kay-burr") made its first appearance in the United States and a whole new round of publicity was launched. Most of the antivirus companies calling me up to talk about it have had this attack squarely in perspective: it was foreseeable because it was possible, and it wont spread very far for a variety of reasons.
It only affects a tiny percentage of mobile phones out there and you have to agree to let it install on the phone. Finally, even if you install it the only real downside is crummy battery life, since it uses the Bluetooth connection excessively, looking for other devices to infect. (Well, theres also the embarrassment of potentially infecting your friends and colleagues phones.
Other PR contacts have referred to "the wide spread distribution of the Cabir Bluetooth cell phone virus" in order to pitch for vendors who provide security software or consulting in this space.
This is simple scaremongering.
Cabir is interesting (more for its use of Bluetooth to seek out and spread to other devices than for the fact it runs on a cell phone), but its not particularly threatening. If I had confidential information in my cell phone that, in the wrong hands, could cause me or my company serious trouble I would think about enhancing the security of it. Of course, for the same reasons I would be worried about forgetting my phone somewhere too, and perhaps thats the more serious threat.
It may have been because of the source code for Cabir was released several months ago, but another one has turned up. Commwarrior (SymbOS/Commwarrior.a to McAfee) affects Nokia Series 60 phones, such as the 3650, 7650, and 6600. How do I know that? I read it on the viruss home page. Its also where I downloaded my own personal copy of the virus, not that I have a Nokia phone on which to run it.