Don’t bring a knife to a gunfight

During 2005 there was an evolutionary leap in the threat landscape as hackers moved toward sophisticated, well-funded attacks for profit. Unfortunately, security professionals and vendors seem to be falling further and further behind. Fighting this evolved threat with legacy anti-virus is like using a wooden shield against a rocket – or bringing a knife to a gunfight. Most organizations are woefully unprotected against this new, more sophisticated threat, and few of them understand how unprotected they really are.

The shift of hacker motives from glory to profit has driven substantial advances in their strategies and technologies – diminishing the effectiveness of legacy anti-virus security solutions dramatically. When glory was their motive, malcode writers wanted to infect hundreds of thousands of systems – but with financial or political motives, such visibility could jeopardize their objectives. Today's hackers want quality footholds, not quantity. They want stealth, not visibility.

When the driving motive is profit, 'total number of infected systems' is a poor indicator of malcode severity. Consider the Israeli espionage trojan that went undetected for nearly two years and in the process stole thousands of confidential documents. Insidious and damaging custom attacks are a rising trend. 'Designer trojans' are running rings around online banking providers. While it didn't make the Top 10 list on any AV website, bank-specific malcode is far more severe than yet another Bagle variant. What keeps us up at night is not Bagle.xyz. Enterprises fear making the same mistakes as ChoicePoint. Governments fear state-sponsored attacks. An AV signature can only be provided after successful fraud because a signature requires a 'patient zero'. With designer attacks, patient zero is the only victim – and a signature is like giving a vaccine to a corpse.

Today's malcode writers have studied legacy defences and have adapted. Two such adaptations were captured in a feature article in the September 2005 issue of Virus Bulletin – short span and serial variant distribution trends (see VB, September 2005, p.9). Short span showed how hackers complete 100 per cent of their infection drive before any AV signatures are released – relegating AV to a 'scan and remove' technology. Serial variant trends showed how releasing several samples extends vulnerability windows and serves to distract and overwhelm AV teams. Add a rootkit and AV may not be able to scan and remove either. Kernel rootkits like ShadowWalker, and VMBRs (virtual machine based rootkits) like SubVirt represent serious threats to any reactive defences. Once they take root, discovery and removal are costly and difficult (see http://www.eweek.com/article2/0,1895,1945808,00.asp). How do you fight an invisible opponent? The enemy has studied our defences, exploited our weaknesses, and changed the nature of the battle.

As Darwin taught us: that which fails to evolve, dies. To turn the tide, the security community needs to make an honest and grounded examination of this 'evolved threat'. The industry needs to look at what can be done to increase its defences, and provide innovative solutions.

The good news is that a handful of virus prevention systems routinely thwart designer attacks and prevent malcode from taking root. These virus prevention systems identify malicious behaviour in any code, new or unknown. The bad news is that the vast majority of us are unaware that effective behavioural technologies exist, and how badly we need them. The worst part is that the majority of the 'trusted security providers' have kept the public in the dark and failed to innovate or keep pace with this evolving threat. We face an escalating danger. Education is the first step – 'knowing is half the battle'.

The bottom line is: don't bring a knife to a gunfight. Your enemy has evolved and is using its latest and greatest arsenal. Are you?

Latest articles:

Bulletin Archive