Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.


09:00 AM
Connect Directly

Russian Hackers Run Record-Breaking Online Ad-Fraud Operation

'Methbot' is a sophisticated cybercrime scheme that has hit major US advertisers and publishing brands and pilfered millions of dollars per day.

[UPDATED 12/21 with response from digital ad industry's Trusted Accountability Group (TAG)]

Cybercriminals out of Russia are behind a newly discovered massive online advertising fraud operation hiding in plain site that steals up to $5 million per day from big-name US advertisers by posing as some 6,000 major US media sites including The Huffington Post, Fortune, ESPN, CBS Sports, and Fox News, and generating fake ad impressions.

Researchers at White Ops recently spotted the so-called "Methbot" operation pilfering anywhere from $3 million to $5 million per day in what they say is the largest and most profitable online ad fraud operation in history. Methbot has been operating for three years under cover by a Russian cybercrime group that White Ops has dubbed "AFK14," with a unique twist: its own internal botnet infrastructure runs and automates the click-fraud rather than the traditional ad fraud model of infecting unsuspecting consumers to do the dirty work.

US advertisers in October alone lost a whopping $17.7 million to the criminal hackers, according to White Ops, and AFK13 made some $10.6 million.

AFK13, which is based in Russia, also employs data centers in Dallas and Amsterdam, to run its botnet via spoofed IP addresses that help them evade blacklists. The cybercrime gang created its own Web browser in order to better hide its tracks, as well as its own HTTP library.

"This is the largest operation ever discovered in digital ad fraud," says Eddie Schwartz, president and COO of White Ops, an ad fraud detection firm, which published its findings on AFK13 and its Methbot infrastructure today. "This one is unique in that they went to the trouble of writing their own browser code … They game everything across the entire value chain" of online advertising, he says.

The Methbot network basically drives video and other ad impressions that appear to be humans clicking on them. But video ad "watching" is actually via its botnet of automated Web browsers of more than a half-million Internet addresses using phony IP registrations posing as large ISPs such as Verizon, Comcast, AT&T, Cox, and CenturyLink.

The botnet generates phony impressions for up to 300 million of these ads daily and sends them via 6,111 Internet domains posing as actual ad inventory on brand-name websites, according to White Ops.  

"Ad companies are losing because they're paying the bill" for phony impressions, White Ops' Schwartz says.

Methbot until recently was able to operate under the radar because the Russian cybergang behind it has apparently studied how to avoid detection, including reverse-engineering and duping ad-fraud measures and spoofing fraud verification data so the advertiser sees Methbot's ad impressions as legit, even though they're phony.

AFK13's Methbot has tallied some 200 million to 300 million phony video-ad impressions daily, making an average of $13.04 per CPM, or around $4 million in phony ad inventory revenue each day.

The Russian hackers even have built the bots to imitate mouse movements and social media login information so they appear to be human-generated activity. "They're making the traffic look like residential humans," Schwartz says.

He says the forged and compromised domains made them appear legit to the advertising exchange services that broker ad space inventory for publishers. The exchanges were fooled into believing they were handing the subsequent ad impressions to the publishers, but that phony yet billable traffic instead went to Methbot.

Methbot Bust

But the fraudulent ad operation ended up exposing itself, thanks to a bug in its homegrown HTTP library. "They used a custom HTTP library that was very buggy and set off a bunch of alerts on our system," says White Ops principal researcher Ryan Castellucci, who is credited with first discovering Methbot. The buggy library was "part of why we noticed that they had started ramping up their traffic," he says.

White Ops declined to identify which advertisers have been victimized by Methbot, but they say it's a who's who in online advertising. The company has been working with law enforcement to unmask the operation.

Mike Zaneis, CEO of the digital advertising industry's Tustworthy Accountability Group (TAG), applauded White Ops' findings. "This massive fraud operation represents a significant threat to the integrity of the ecosystem, and it shows why TAG's work is so vital in bringing the digital advertising industry together to share information, adopt rigorous standards, validate best practices, and increase transparency," Zaneis said in a statement.

"Within 24 hours of our notification by White Ops, TAG was able to alert 130 fraud compliance officers at the largest and most influential digital advertising companies and bring the vast majority of those anti-fraud leaders together to learn details of the attack and determine the appropriate action for their companies to take," he said.

Here's how big Methbot stacks up to previous click-fraud campaigns money-wise: ZeroAccess took in about  $900,000 per day; the Chameleon botnet, up to $200,000 per day; and HummingBad, up to $10,000 per day, according to White Ops' report.

Related Content:

Kelly Jackson Higgins is the Executive Editor of Dark Reading. She is an award-winning veteran technology and business journalist with more than two decades of experience in reporting and editing for various publications, including Network Computing, Secure Enterprise ... View Full Bio

Recommended Reading:

Comment  | 
Print  | 
More Insights
Newest First  |  Oldest First  |  Threaded View
User Rank: Apprentice
8/24/2017 | 5:46:28 AM
Re: logo
yes it was, saw it too
User Rank: Apprentice
12/26/2016 | 1:27:41 AM
This is such a great resource that you are providing and you give it away for free.    
pengalaman melancong
User Rank: Moderator
12/23/2016 | 9:57:30 AM
Re: Stats
What I don't understand is this:

Why the pay models for advertisement are not based on sale?  An advertisement that was seen by one hundred and netted $1,000 in sale should be paid more than an advertisement that was seen by one million people but netted only $100.  Prior to digital age, it was difficult to track what form of advertisements generated the final sale.  Now that we have all those fancy algorithms and consumer behavioral collections, it should be much easier to track the relationships between the advertisement and sale.  So why does the model of greater audience for bigger advertising dollars still persist, which is pretty much the main driver of the ad-fraud operations?
Joe Stanganelli
Joe Stanganelli,
User Rank: Ninja
12/22/2016 | 7:05:09 PM
Re: A Hollow Economy
Someone I'm close with is a digital marketing exec.  Combined with my own consulting in the space, when we talk, we're flabbergasted with how little people know about PPC and other online ad spending -- and how marketing charlatans are able to make so much money off of them for doing so little (often even being counterproductive)!
Joe Stanganelli
Joe Stanganelli,
User Rank: Ninja
12/22/2016 | 6:59:13 PM
I remember seeing a stat a year or two ago that about two thirds of all online ad impressions/clicks were bots -- and that Google actually knew this but was doing pretty much nothing about this (thus sparking complaints).

That was then.  It's clear that things haven't changed for the better much -- and may have even gotten worse.  Good to see one of these operations exposed to some degree.
User Rank: Apprentice
12/21/2016 | 3:29:54 PM
Analyzing the data yourself is best
If you are paying for digital advertising and monitor your analytics it will be clear in the data if you are being exploited.  In looking at some web media buys we saw a majority of clicks were from a specific line of Nokia Windows mobile devices hitting our pay per click campaign.  The SEO / PPC company you hire probably will over look or ignore any of this - I mean how many people really use a windows mobile device these days!  Fighting the ad network (Google or others) can get you a refund if you have a legitimate arguement. 
User Rank: Moderator
12/21/2016 | 6:56:09 AM
Re: A Hollow Economy
'Hollow Economy' is a spot-on term. Well said.
User Rank: Ninja
12/20/2016 | 4:10:39 PM
A Hollow Economy
I've been doing automated web testing for years; almost 20 years, in fact.  What surprises me the most about this article is not the "hack" itself (which I can't believe some variation on this hasn't been done already, essentially since after the moment this ad pay model appeared), but that this model of ad pay still is being used.  The very fact such huge daily amounts of revenue could be pulled in even by legit means seems ludicrous to me.  Perhaps I feel that way because I'm not out there taking advantage of this model (legally, of course), but more it just disturbs me that we continue to open ourselves up to 1) hacks of this type and 2) bad business practices.  The invisible economy - hollow economy - persists and can only lead to bad things down the road for everyone, much as the dot-com bubble burting taught us.
User Rank: Apprentice
12/20/2016 | 12:43:53 PM
very clever operations by a knowledgeable group
they knew the ins and outs of browser technology, ip technology and ad tech... and probably used hundreds of publisher accounts not to get caught, after all who makes 3 million a day by serving video ads? thats a lot of money.
US Formally Attributes SolarWinds Attack to Russian Intelligence Agency
Jai Vijayan, Contributing Writer,  4/15/2021
Dependency Problems Increase for Open Source Components
Robert Lemos, Contributing Writer,  4/14/2021
FBI Operation Remotely Removes Web Shells From Exchange Servers
Kelly Sheridan, Staff Editor, Dark Reading,  4/14/2021
Register for Dark Reading Newsletters
White Papers
Current Issue
2021 Top Enterprise IT Trends
We've identified the key trends that are poised to impact the IT landscape in 2021. Find out why they're important and how they will affect you today!
Flash Poll
How Enterprises are Developing Secure Applications
How Enterprises are Developing Secure Applications
Recent breaches of third-party apps are driving many organizations to think harder about the security of their off-the-shelf software as they continue to move left in secure software development practices.
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
PUBLISHED: 2021-04-20
An unsafe deserialization vulnerability in Bridgecrew Checkov by Prisma Cloud allows arbitrary code execution when processing a malicious terraform file. This issue impacts Checkov 2.0 versions earlier than Checkov 2.0.26. Checkov 1.0 versions are not impacted.
PUBLISHED: 2021-04-20
An information exposure through log file vulnerability exists in Palo Alto Networks PAN-OS software where secrets in PAN-OS XML API requests are logged in cleartext to the web server logs when the API is used incorrectly. This vulnerability applies only to PAN-OS appliances that are configured to us...
PUBLISHED: 2021-04-20
An information exposure through log file vulnerability exists in Palo Alto Networks PAN-OS software where the connection details for a scheduled configuration export are logged in system logs. Logged information includes the cleartext username, password, and IP address used to export the PAN-OS conf...
PUBLISHED: 2021-04-20
A denial-of-service (DoS) vulnerability in Palo Alto Networks GlobalProtect app on Windows systems allows a limited Windows user to send specifically-crafted input to the GlobalProtect app that results in a Windows blue screen of death (BSOD) error. This issue impacts: GlobalProtect app 5.1 versions...
PUBLISHED: 2021-04-19
An out-of-bounds (OOB) memory access flaw was found in fs/f2fs/node.c in the f2fs module in the Linux kernel in versions before 5.12.0-rc4. A bounds check failure allows a local attacker to gain access to out-of-bounds memory leading to a system crash or a leak of internal kernel information. The hi...