Sunday, April 21, 2013

Bing denies results from the recent AV-Test search engine study

A few weeks ago, AV-Test conducted a study on Google and Bing, and how many malware infected results they deliver. The result of the study showed that Bing served up 5x more infected results than Google did. However, Microsoft says that the study is not true, for a simple reason - the study is flawed.

Last Friday, a blog post by David Felstead, Senior Development Lead at Bing said that the study was flawed because AV-Test used a Bing API to conduct searches instead of using the website, which displays warnings about known to be infected or suspicious websites. Bing's API bypasses the warning system and warning labels do not show.

Felstead goes on to say that Bing does block suspicious websites, by disabling the link on the search results. However, why not block them completely? The reason they don't is because sometimes websites get hacked and deliver malware, but get fixed.
The top result is a suspicious website. When you try to click on the link, you'll see the warning box to the right. If a user still wants to continue, they can still do so.
Here's the results from Google:
Aside from the search ads, you can see that the malicious result is still at the top of Google's search results, but there is no warning about the danger of visiting the page. This effectively shows that Bing can block malicious websites.
However, it is good to keep note that this isn't a stab to say that Bing is good and Google is bad. "No, it means this is a highly complex problem that all engines are constantly working to solve.  No engine will be perfect 100% of the time but we all work every day on detecting the latest threats from the bad guys and updating our engines to keep customers safe," writes Felstead.

Yandex, another search engine, is also questioning this study. They question how AV-Test determines a result to be malicious or not, among other things. Search Engine Land wrote a great article about this, so refer to them.

So, what do you think?