Why Can't Google Stop Malware Ads on Adwords?

People make much of technical matters in security, but the most important force behind malware is social engineering, not some vulnerability or bad design. The current hot malware is a textbook case of social engineering and an aggressive marketing campaign.

You must have seen them by now: ads for "Antivirus XP 2008" or some variant of that name. In fact, here are some of the newer names used for essentially the same attack. The program is not malicious in most conventional senses: It doesn't spread itself to other systems surreptitiously, it doesn't steal passwords or anything like that. Instead it claims to be a security program. It performs a fake scan of the system and then claims to find numerous threats which the user can remove by buying a full license to the product (for $49.95).

I've seen spam pushing it. A series of hijacks of ad networks late in 2007 looks like it was the same malware or a variant of it. The most popular way to distribute it these days, I'm told by Sunbelt Software, is through affiliate-based malware sites. On such sites the user clicks to download something, like a codec for a video, and gets a variety of malware, including this one.

But the most interesting way this threat has been spread has been is through advertising, most infamously through Google sponsored links.

This has been going on for a while and it seems as if Google can't stop them. It's true that the scammers have been using a wide variety of product and domain names; for some more reports of new domains and IPs implicated in the threat see here and here. But if that's all it takes then the bar for getting past Google's defenses against such products advertising through AdWords and AdSense are thin indeed.

The question then becomes: Does Google have any defenses against such products? Or is the worst that it comes to is that someone might complain?

There is a prohibited uses clause in the AdSense Terms and Conditions page. The only page I can find labeled as AdWords Terms and Conditions is blank. But that's just an agreement, and no indication that Google actually checks sites associated with ads submitted, or even has a way to check them after the fact or a policy for when to do so.

I asked Google about this problem and the general problem of malicious sites advertising through it. A Google spokesperson responded, "Google actively works to detect and remove sites that serve malware in both our ad network and in our search results. We have manual and automated processes to do this. We have canceled advertising accounts that display ads redirecting users to malicious sites or that advertise a product violating our software principles. For more information on our general software principles, please check out http://www.google.com/corporate/software_principles.html."

I should add that I'm sure they get reports from the outside, from services like Grisoft's AVG LinkScanner of malicious sites. But it sounds like it's all reactive; clearly malware sites can and do get through and, before they get caught, they're there for people to click on. And starting another AdSense/AdWords account, especially if you're not really interested in being paid for the clicks, is not hard.

Google is trying to improve things, and certainly it's in Google's interest to do so. The company added the following remarks:

We've set up a number of automated systems to scour our index for potentially dangerous sites, and we add a label to those that appear to be a vehicle for malware. If you're searching on Google and click on a link that we've flagged, a warning page will appear in search results before you move forward.  We flag sites that serve malware directly or that contain ads carrying malicious content.  Some of the URLs we add to the blacklist are ads.  It's important to note that not all browsers check all the resources on a page, such as ads, against the malware blacklist we provide. As always, we are constantly working to increase our coverage of malicious content on the Web.

Some time ago I remember quite a bit of controversy over whether users could discern the difference between paid ads and "organic" links in a Google page. I'm sure there are still plenty of people who can't tell the difference, and for those who can some number of them probably assume that the fact that Google is putting up a "Sponsored Link" means that Google has done some sort of checking on it and is somehow vouching for it.

This is, of course, completely untrue, just as the sites that host Google ad network advertisements have no responsibility for the Google ads on them. There's no good advice to give users here other than to be suspicious of Google ads just as they would of any potentially unfriendly link. And run a product or service, like LinkScanner, which checks browser links against a list of known bad ones. Yes, you can't even let your guard down on Google.

Security Center Editor Larry Seltzer has worked in and written about the computer industry since 1983.

For insights on security coverage around the Web, take a look at eWEEK.com Security Center Editor Larry Seltzer's blog Cheap Hack.