Dataium Settles Browser History Sniffing Charges
The car buyer tracking firm was accused of using JavaScript to illegally identify websites visited by 181,000 named consumers, and selling harvested information.
A data broker accused of illegally accessing consumers' website browsing histories has settled related charges filed against it by New Jersey state's attorney general.
The terms of the agreement with Nashville, Tenn.-based Dataium includes a one-time $99,000 fine, payable to the state over the next two years. Meanwhile, the company will be liable to pay a suspended sum of $301,000 to the state if it fails to honor the terms of its settlement agreement at any time over the next five years.
{image 1}
"Dataium allegedly used software code to track the websites visited by consumers without their knowledge or consent. The company also allegedly transferred the personal information of 400,000 consumers to one of the largest data brokers in the world," said New Jersey Division of Law director Christopher S. Porrino. "At the very least, Dataium should have notified consumers and disclosed their data transfer practices."
The settlement agreement also stipulated that by December 15, Dataium must detail on its website exactly which types of consumer information it's collecting; how that information is being used, shared, or sold; and how consumers can restrict or opt out of that information collection. Within 90 days, Dataium must also create a privacy program, designate a single employee to oversee that program, and undergo regular audits of that program for the next five years.
According to its website, Dataium is "the world's largest compiler of online automotive shopping behavior" and monitors the behavior of 20 million would-be car buyers every month. The company has said that 10,000 different automotive sites employ its tracking technology.
[ Is your TV spying on you? Read LG Admits Smart TVs Spied On Users. ]
"The automotive websites that work with Dataium install code on their site that log everything an individual visitor might do on a website, from what pages they visit, to where they click on the page, to even any contact information a visitor might submit on a contact form or select from a drop-down menu," reads an overview of Dataium's tracking technology published by privacy software firm Abine. "Dataium also installs tracking cookies on a visitor's computer to follow where they browse and specifically track what other automotive websites a user visits while they do their automotive shopping." Some of this data is then sold -- in non-anonymized form -- to larger data brokers.
New Jersey state consumer protection officials began investigating Dataium over its alleged "history sniffing" practices -- referring to a technique used to surreptitiously identify which sites a user has previously visited -- following The Wall Street Journal's report in December 2012 that Dataium was using JavaScript code -- "buried behind at least four layers of obfuscation" -- to gather history-sniffing data and then tying that data to people's names.
"As in numerous other cases regarding data privacy over these past several years, this media scrutiny led to deeper scrutiny by public officials," said ZwillGen attorney Ken Dreifach in a blog post.
According to one 2010 study of privacy-violating information flows in JavaScript web applications written by four researchers at the University of California at San Diego, a history-sniffing attack involves using JavaScript to study whether a browser regards another link listed on the page as having been visited. "A malicious site (say, attacker.com) can learn whether a user has visited a specific URL (say, bankofamerica.com), merely by inducing the user to visit attacker.com," said the researchers. "To this end, the attack uses the fact that browsers display links differently depending on whether or not their target has been visited."
Ultimately, the state accused Dataium of running six related "internal" trials between November 2010 and November 2012. According to the state's consent order, in just one of Dataium's history-sniffing trials, in November 2012, the company tracked approximately 181,080 user visits to various car dealerships' websites, popular search engines, and news articles -- about 100 sites in total. That followed a September 2012 business agreement between Dataium and data broker Acxiom. "Dataium entered into a Data Supplier Test Agreement to transfer the personal information of 400,000 consumers to Acxiom, one of the world's largest data analytics companies, to determine the value of marrying online behavior from Dataium with offline behavior from Acxiom," read the consent order.
The state said that Dataium sold the information it had collected on the 181,000 consumers to Acxiom for $2,500. That collected information included a person's name, email address, a unique ID number assigned by Dataium, the first and last date that the consumer was active, and a number of details relating to different types of vehicles consumers viewed along with their apparent preference for each vehicle.
In the consent decree, the state accused Dataium of illegally transferring that information to Acxiom without the knowledge or express consent of the consumers. Dataium denied that allegation.
Independent privacy and security expert Ashkan Soltani said via Twitter that the state's settlement agreement "contains a rare glimpse into Acxiom's data supplier agreements."
You can use distributed databases without putting your company's crown jewels at risk. Here's how. Also in the Data Scatter issue of InformationWeek: A wild-card team member with a different skill set can help provide an outside perspective that might turn big data into business innovation. (Free registration required.)
About the Author
You May Also Like
Unleashing AI to Assess Cyber Security Risk
Nov 12, 2024Securing Tomorrow, Today: How to Navigate Zero Trust
Nov 13, 2024The State of Attack Surface Management (ASM), Featuring Forrester
Nov 15, 2024Applying the Principle of Least Privilege to the Cloud
Nov 18, 2024The Right Way to Use Artificial Intelligence and Machine Learning in Incident Response
Nov 20, 2024