Malvertising: The story behind the story

A piece of paper tacked to a cork board, that reads "stop guessing start knowing"

Security firms make mountains out of molehills

Malware alert! Malware alert! It seems every time you turn around there’s a news story or report exposing the presence of malware in the online and mobile advertising ecosystem. The vector, exploit kit or function may change, but the story is the same—some industry expert uncovers new ad-based malware or malvertising and the media sounds the alarm. Preying on cyber-related anxieties, these stories typically present an exaggerated synopsis of the situation and focus on a single instance, spotlight one industry provider, and don’t offer actionable information for the reader. As a result, these provocative articles often make mountains out of molehills and end up missing the real story: Why does the industry expert believe this particular malware incident is news?

Keeping it real

Malware serves as an umbrella term for any intrusive software program with malicious or hostile intent, and covers a variety of forms including viruses, Trojans, and worms. Diagnosing malware provides critical insight into identifying current system vulnerabilities and mitigating future compromises and the classic approach used by traditional security researchers requires the collection of malware samples and days of analysis by experts.

Ad-related malware behaves differently from other forms of malware and requires a distinct approach. Anyone that truly understands the advertising ecosystem recognizes that ad-based malware delivers through a publisher website for a very brief time period, typically for an hour or less, before it terminates and moves on in a mutated form to infect hundreds of other sites. In addition, the infected ad must first render on a browser before it deploys—automatically or through site visitor action—and there’s no guarantee that it will impact every browser or deploy every time rendered.

For these reasons, it’s misleading to report on one malvertising incident captured on one site. In addition, it’s irresponsible to call out a publisher for something that cannot be replicated, and these reports cause unnecessary panic among advertisers, ad networks, exchanges and publishers who spend countless resources addressing a malware event that no longer exists.

Diagnosing the motivation

Publishing incident-specific ad-based malware reports provides very little useful information and does very little to eliminate malvertising from the advertising ecosystem. Yet, this reporting persists for two primary reasons—extortion or publicity.

Known as “White Hat Ransomware”, disreputable security analysts mine websites for malvertising incidents and present the findings to the site/publisher hosting the bad ad. They offer to sell the vector information so the publisher can shut down the infection, with the understanding that the malware incident could be publicly released should the publisher choose to not pay. Usually perpetrated by obscure individuals or groups, this type of extortion proves very lucrative as many publishers purchase the information in order to avoid the time-consuming fallout of negative publicity.

The more reputable network, endpoint and intelligence security firms try to extend their traditional malware analysis skill set to malvertising and digital content. However, it doesn’t work. Effective analyses requires continuous, real-time monitoring of the advertising environment from the browser or consumer point of view which requires scanning active ad placements using simulated users set up with the exact geographic and behavioral profiles that the ad is targeting—something that can’t be accurately replicated after the fact. In addition, the ever-shifting nature of malvertising means that capturing a screen shot of an incident found on a single site is misguided—if it exists on one site, it exists on hundreds or thousands of other publisher sites and ad networks—and the post-incident analysis offers no valuable benefit to the consumers already exposed. By publishing malvertising-related reports about something that happened days, weeks or months ago, these firms unleash chaos in the ad tech industry as the publisher and its partners attempt to locate a vector that no longer exists.

Protecting the advertising ecosystem

Malware in the ad tech industry is not news. Admittedly, the ad tech industry plays a central role in the propagation of malware in the online and mobile advertising ecosystem, however, this fact is not ignored by responsible industry players who fiercely combat it every day. From establishing working groups to creating “good ad” certifications to performing extensive due diligence on buyer clients, the industry works hard to tackle the presence of malware. In fact, many of largest, most-visited websites actively scan their advertisements to identify and remove anomalous vectors before they morph and become overt malware drops. Unfortunately, a few ad-based malware vectors get through, but that number is minuscule in comparison to the billions of ads successfully rendered every day.

In effect, malvertising isn’t a new trend. In fact, it emerged shortly after the birth of banner ads 20+ years ago. What’s new is that traditional security companies are finally realizing that digital properties—websites and mobile apps—can be compromised. If you want to know how malvertising really works, ask The Media Trust. We’ve been detecting malware in the online and mobile environment for close to a decade, not the past few months.