Antivirus Software Not So Useless as People Say: Report

While companies continue to call out the antivirus industry for the malware it misses, NSS Labs says that most criticism is based on misconceptions.


Antivirus software has gotten a bad rap, even though the programs continue to help protect firms and raise the bar for attackers, according to a report released on March 11 by research firm NSS Labs.

The report finds that antivirus software has largely been criticized based on two-decades-old misconceptions. The software running on users' computers does far more than matching patterns, or signatures, for known threats. Instead, it uses a variety of technologies—from firewalls and host intrusion detection to behavioral heuristics and anomaly detection—to find what is likely to be malicious software, or malware, Randy Abrams, research director at NSS Labs, told eWEEK.

Today's anti-malware software—the term "antivirus" is another hold out from the days of a less complicated threat—does the equivalent of the credit check required by a bank, looking at a variety of factors to assess risk, he said.

"If you want a loan at a bank, you have to have a reputation that they call a credit score," Abrams said. "If you have a low credit score, it does not mean you wouldn't pay the bank back; it just means that you are a greater risk."

Despite increasingly nuanced detection strategies, the antivirus industry has become the punching bag of the security industry, mainly because it is the last line of defense against the compromise of a computer that could, and most often does, lead to a major breach. When a computer is infected with malware, users do not blame the firewall or the network intrusion-detection system—they criticize the software that protects the endpoint, Abrams said.

When Chinese hackers infiltrated The New York Times, for example, the media latched onto the fact that only a single program used by the attackers was detected by the Symantec anti-malware. In fact, only 24 percent of the malware used in similar incidents has historically been detected by any endpoint-security solution, according to a Mandiant report.

While Abrams may sound like an apologist for the endpoint-protection software industry, he is not. The industry has evolved better defenses but still has not done enough, Abrams said. Because attackers have time to test against the most major antivirus products and find ways to circumvent their defenses, no software program can protect the endpoint indefinitely.

Yet, rather than resign themselves to that fact, anti-malware firms need to evolve and build a more holistic defense against attackers, not only detecting and blocking bad software, but detecting signs of, and helping users remediate, a compromise before the breach gets worse, he said.

"When some other companies detect a breach, the antivirus firms look bad, but when their software detects the breach, then they look like they are doing their job," Abrams said. "By taking on both tasks, they are creating a much better system of security."

Robert Lemos

Robert Lemos

Robert Lemos is an award-winning freelance journalist who has covered information security, cybercrime and technology's impact on society for almost two decades. A former research engineer, he's...