Ad Injection at Scale: Assessing Deceptive Advertisement Modifications

Kurt Thomas, Elie Bursztein, Chris Grier, Grant Ho, Nav Jagpal, Alexandros Kapravelos, Damon McCoy, Antonio Nappa, Vern Paxson, Paul Pearce, Niels Provos, Moheeb Abu RajabIEEE Symposium on Security and Privacy 2015 · OpenReview

Abstract

Today, web injection manifests in many forms, but fundamentally occurs when malicious and unwanted actors tamper directly with browser sessions for their own profit. In this work we illuminate the scope and negative impact of one of these forms, ad injection, in which users have ads imposed on them in addition to, or different from, those that websites originally sent them. We find that ad injection has entrenched itself as a cross-browser monetization platform impacting more than 5% of unique daily IP addresses accessing Google—tens of millions of users around the globe. Injected ads arrive on a client's machine through multiple vectors: our measurements identify 50,870 Chrome extensions and 34,407 Windows binaries, 38% and 17% of which are explicitly malicious. A small number of software developers support the vast majority of these injectors who in turn syndicate from the larger ad ecosystem. We have contacted the Chrome Web Store and the advertisers targeted by ad injectors to alert each of the deceptive practices involved.

Introduction

Ad injection represents a significant yet often overlooked threat in the modern web ecosystem. By tampering with the Document Object Model (DOM) of a user's browser, malicious or unwanted software can insert advertisements that were never intended by the website owner. This practice not only degrades the user experience through increased latency and intrusive UI elements but also siphons revenue from legitimate publishers and advertisers.

While traditional malware might focus on data exfiltration or system disruption, ad injection occupies a lucrative "gray area" of Potentially Unwanted Programs (PUPs). Because the modification occurs on the client side, it is frequently invisible to website administrators, leading users to believe the host site is responsible for the deceptive content. This research provides the first large-scale look at the scope and financial underpinnings of this ecosystem.

Methodology and Detection Pipeline

To quantify the scale of this problem, the researchers developed a multi-staged detection pipeline. The core of this system is a client-side DOM scanner deployed on a subset of Google properties. This scanner compares the live page structure against a whitelist of expected elements to identify rogue scripts, iframes, and links.

The researchers used "trigger pages" from high-traffic sites like Google, Amazon, and Walmart to observe the behavior of suspected injectors in a controlled environment. By automatically interacting with injected ads, they were able to map the "clickchains"—the series of redirects leading from the injector to the final advertiser.

Overview of ad injection across major platforms: Google, Amazon, and Walmart.

The Ad Injection Ecosystem

The study analyzed a massive dataset comprising over 100 million DOM reports, 1 million Chrome extensions, and 25 million Windows binaries. The findings reveal a highly structured hierarchy consisting of injection libraries, affiliates, and intermediaries. A small number of developers (e.g., Superfish, Jollywallet) provide the core injection technology, while thousands of distribution partners bundle these libraries to monetize their user base.

A detailed example of injected 'Hot Deals' and 'Recently Bought' overlays on an Amazon search results page.

Injected traffic is often funneled through a few key ad-tech "bottlenecks" like ShopZilla or PriceGrabber before reaching the final advertiser. These intermediaries represent the most effective points for intervention to disrupt the revenue chain.

Key Results and Impact

The prevalence of ad injection is staggering, affecting 5.5% of unique daily IPs visiting Google properties. The researchers identified over 80,000 unique software components (extensions and binaries) involved in these activities. Notably, 38% of the identified extensions and 17% of the binaries were flagged as explicitly malicious by security systems.

The financial scale is equally impressive; for example, Superfish.com was found to have generated over $35 million in gross revenue in 2013 alone. This research led to immediate real-world impact, resulting in the removal of 192 deceptive extensions from the Chrome Web Store, protecting approximately 14 million users.

Conclusion and Future Outlook

Ad injection is a pervasive threat that exploits the complexity of the modern ad-tech stack. The authors advocate for stronger browser protections against unauthorized extension "side-loading" and the widespread adoption of HTTPS/HSTS to mitigate network-level injection. Addressing the financial incentives by targeting bottleneck intermediaries remains a primary strategy for securing the web against these deceptive modifications.

Summary updated

Abstract

Today, web injection manifests in many forms, but fundamentally occurs when malicious and unwanted actors tamper directly with browser sessions for their own profit. In this work we illuminate the scope and negative impact of one of these forms, ad injection, in which users have ads imposed on them in addition to, or different from, those that websites originally sent them. We develop a multi-staged pipeline that identifies ad injection in the wild and captures its distribution and revenue chains. We find that ad injection has entrenched itself as a cross-browser monetization platform impacting more than 5% of unique daily IP addresses accessing Google -- tens of millions of users around the globe. Injected ads arrive on a client's machine through multiple vectors: our measurements identify 50,870 Chrome extensions and 34,407 Windows binaries, 38% and 17% of which are explicitly malicious. A small number of software developers support the vast majority of these injectors who in turn syndicate from the larger ad ecosystem. We have contacted the Chrome Web Store and the advertisers targeted by ad injectors to alert each of the deceptive practices involved.
BibTeX
@inproceedings{thomas2015ad,
  author = {Kurt Thomas and Elie Bursztein and Chris Grier and Grant Ho and Nav Jagpal and Alexandros Kapravelos and Damon McCoy and Antonio Nappa and Vern Paxson and Paul Pearce and Niels Provos and Moheeb Abu Rajab},
  title = {{Ad Injection at Scale: Assessing Deceptive Advertisement Modifications}},
  booktitle = {IEEE Symposium on Security and Privacy},
  year = {2015},
  url = {https://openreview.net/forum?id=uQMGvirZ3Pa},
}
Figure