Malvertisements Hit 3 Million Impressions per Day at End of 2010

There were over 3 million impressions on malicious Web advertisements in the final quarter of 2010, as a result of poor vetting by ad networks and social media.

The number of malicious Web advertisements jumped sharply in the fourth quarter of 2010, according to a report from security firm Dasient.

"Malvertising," or advertisements containing malware, was on the rise, with more than 3 million impressions per day in the fourth quarter, double what was found in the third quarter, Dasient said in its 2010 Malware Report, released Mar. 7.

Attackers were increasingly sneaking malicious code and links to malware-laden Websites within malicious ads that legitimate advertising networks were distributing to a large number of Websites. Security experts predict that these ads would increase in 2011 as a very easy and lucrative attack vector.

Part of the increase in Dasient's report may be attributed to the fact that the company decided to include remnant ad networks in the survey for the first time. Remnant networks sell empty advertising slots for a low rate so that other advertising networks have another pool of ads to draw upon instead of relying on house ads to fill empty inventory, Dasient said. These third-party ad networks aggregate advertisements and generally have very low ad rates, so they are less likely to vet such advertisements carefully, the report said.

"With the addition of more remnant ad networks in our telemetry, we believe that we are more accurately reflecting the current state of malvertsing," the report stated.

Malvertising can wreak havoc on unsuspecting Website visitors without going through the trouble of hacking the actual Website. Such was the case recently when visitors to the London Stock Exchange Website were hit by rogue anti-virus malware served up by a malicious ad that appeared on the site.

As part of the research, Dasient researchers conducted several "benign" experiments to see how malware was being spread. Researchers posted an ad whose click-through links pointed to a Website that automatically triggered a drive-by-download, Dasient said. The file being downloaded was safe, the researchers said. The benign ad had the headline, "Click for a security test," was linked to "" and opened up the Windows calculator if the computer was vulnerable. Researchers said various social-networking sites kept the ad up for more than three weeks.

Social networks are also not screening their ad landing pages with automated systems that look for drive-by downloads of malware, said Dasient.

Firms that serve advertisements need to do a better job vetting the content of the images they distribute for malicious code and detecting Web-based attacks, including malicious ads, when they appear, said Neil Daswani, Dasient's CTO.

The researchers also tested how easy it was to spread malware via malicious links on social-network sites. More than 80 percent of the social-networking sites allowed links that were on Google's Safe Browsing list to pass, and all the sites allowed links that triggered a drive-by download, the researchers said. Most social-media networks do not check links in UGC (user-generated content) such as posts and comments to determine if they contain or lead to malware threats, the report found.

It was clear that infections could occur relatively easily via UGC interactions and advertisements, according to researchers.

More than a million Websites were infected with malware in the fourth quarter of 2010, nearly double those of the last quarter 2009, according to the report.