There has recently been an unfettered flow of advertising for malicious software on Google's AdWords networks. How come Google can't stop the malware?
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.
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,
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
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
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
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.
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.