What if the “C” in “CTV” stood not for “connected” but for “counterfeit”?
That is the growing fear of many ad buyers, as the scale of the digital ad fraud challenge crashes on to the medium’s shores.
In this video interview with Beet.TV, Tamer Hassan, Co-Founder & CEO, HUMAN Security, a cybersecurity firm, lifts the lid on the practices hitting connected TV.
Falsified bundle IDs
In September, HUMAN’s team identified a fraud scheme it named Scylla, wherein 29 Android apps were pretending to be more than 6,000 CTV-based video apps, in order to bring in higher ad CPMs.
It was the third iteration of a widespread fraud scheme dubbed Poseidon – but this was the first to have manifested on iOS.
According to HUMAN’s investigation:
“They do this by inserting a different “bundle ID” or “app ID” (think driver’s license) in the code. …
“The apps in the Scylla operation are instructed which bundle ID to use by a remote command-and-control (C2) server, which tells the app which bundle ID to dynamically insert in the code for that specific impression opportunity, all in a matter of milliseconds.”
BREAKING: HUMAN’s Satori Threat Intelligence & Research team has discovered an operation codenamed Scylla, which targets SDKs within mobile apps that have been downloaded over 13 million times. Learn more: https://t.co/udSO29aABh #adfraud #cyberattacks #fraud #KnowWhosReal
— HUMAN (@SecureWithHUMAN) September 23, 2022
Fake TV views
Hassan tells Beet.TV: “There is a range of fraud models that happen, but some of the more common ones are impersonation and amplification.
“It’s very costly to infect a million TVs, but if you’ve infected a million phones or a million PCs and you can manipulate the calls to make it look like a TV, that’s what you’ll do.”
Smart mobile fraud
Hassan says the problem is multi-pronged.
“There’s a lot of ways to launder human eyeballs and attention and, and it doesn’t have to be simply setting up a website and doing fraud,” he says.
“If you’re anywhere in the middle, you can start to amplify. You can start to essentially make counterfeit CTV, as an example, which is one of the more prolific fraud models here.”
The Posiedon/Scylla variant was particularly smart,” HUMAN’s CEO adds.
“It was a batch of Android applications installed on millions of users,” Hassan says. “They had logic built in that would even detect when your phone was out of sight in your pocket, maybe walking or face-down, and then it would start running ads to deliver.