Personally, I wouldn't call a program that hooks itself into the kernel and often runs several services at system/administrator level "secure". Not all security programs use Kernel hooks but many do...
...If a piece of software is hooking into the kernel, it doesn't matter what you do, because the potential for exploitation forfeits any benefit given by whatever AV program is running.
There are plenty of hooks into any OS kernel but to get access to them you have to elevate privilege in turn. AV products are hardened against that and quite effectively so. Very few pieces of malware will have a negative effect on AV systems - the engines are patched as soon as exploits are discovered. It's a very low risk.
I always prefer preventative measures of stopping the bad stuff getting in, and I'm just talking simple stuff like ACLs on sensitive parts of the system, strengthening IE settings, etc. There was a guy on another forum I used to frequent who always said "Security does not come with an installer", and to be honest, I couldn't sum it up better. You can spend all day long gumming up a system with anti this and anti that, but at the end of the day it's just one more application, one more service, one more potential avenue for attack. As a Security professional, I would hope that you can appreciate that.
Prevention is always better than cure and the statement that security doesn't come with an installer is absolutely correct BUT it doesn't mean that installed software isn't a part of the arsenal. It means that you have to start with policy and culture - a security mindset if you like. After that comes the technical and physical controls each of which is subject to proper and thorough risk analysis.
At the perimeter you have your firewalls performing at proxy, stateful and port level. Within the LAN you apply ACLs to switches, routers and other appliances and use VLANs to segregate traffic as appropriate. You may even have secondary internal firewalls to isolate critical systems. You deactivate unused ports as well. On each host device you have protective measures including least privilege, malware protection (including AV), endpoint control and encryption if necessary. You may even have a local software firewall - advisable on any remote or mobile computing device. All of this is centrally managed and tamper proof. Finally you have a timely patch management policy. There are other bits and pieces but that's the core of it.
Properly managed all of this enhances security without increasing risk and without consuming excessive resources. Security is supposed to add value not impede and proper implementation will achieve this.
I am not a Security professional on paper, but I do work at the front line of tech support and everyday I have to clean up the mess left behind that should be taken care of by AV programs, if they are half as good as what they are cracked up to be. The reality of it is, by and large they aren't worth the paper they're written on - companies pay an absolute fortune for so called "protection", and yet whenever the latest malware strikes, technicians are still running around sorting it. I've been completely antivirus free on all my computers since around 2007 now, and have seen absolutely no problems. Other people I know who aren't quite so Internet savvy, have AV programs installed, and I am still around their house every 6 months sorting it.
If your corporate AV systems aren't protecting you, then you need to do several things. 1. Embed a security culture with your users. ISO27002 is a good place to start. Educate them and bind them with policy. 2. Examine your system configuration and AV performance. If the systems aren't catching the malware then either they're badly implemented or they aren't
up to the job and you need different ones. They do work if they're competent and correctly configured.
We have around 800 hosts over 60 sites and the only major problem we've had is the conficker one - introduced on a rogue USB device. The systems caught it, alerted us, blocked its spread and cleaned it with very little technician intervention.
Taking all of this into account, I hope you can understand where I am coming from when I say that Antivirus is snake oil, as I personally have never seen evidence to suggest otherwise, in work or at home. Perhaps I should have said, I believe that Antivirus is snake oil... but I'm not going to say that you are "wrong" or that I am "right".
You may believe it but empirical evidence indicates otherwise. In any event you understand and manage your own risks and that's fine - they're your own risks.
However, to suggest to others who may not have the same level of understanding that they don't need a major security component and to denigrate that component when every area of research and the security standards themselves indicate that they are both effective and an important part of the system defence is irresponsible I'm afraid.
Apologies to the OP for the thread hijack
