text
stringlengths
0
118
would rise significantly due to bank reimbursement demands, regulatory
fines, and direct customer service costs.8 About 90 lawsuits were filed,
leading to massive lawyer bills.9
What made this all the more unnerving for Target is that it had devoted
quite a lot of time and resources to its information security. Target had more
than 300 information security staff members. The company had maintained
a large security operations center in Minneapolis, Minnesota, and had a
team of security specialists in Bangalore that monitored its computer
network 24/7. In May 2013—just six months before the hack—Target had
implemented expensive and sophisticated malware detection software from
FireEye.10
With all this security—an investment of millions of dollars, state-of-the-
art security software, hundreds of security personnel, and round-the-clock
monitoring—how did Target fail?
A common narrative told to the public is that this entire debacle could be
traced to just one person who let the hackers slip in. In caper movies, the
criminals often have an inside guy who leaves the doors open. But the
person who let the hackers into Target wasn’t even a Target employee and
wasn’t bent on mischief. The person worked for Fazio Mechanical, a
Pennsylvania-based HVAC company, a third-party vendor hired by Target.
The Fazio employee fell for a phishing trick and opened an attachment in a
fraudulent email the hackers had sent to him. Hidden in the email
attachment lurked the Citadel Trojan horse—a malicious software program
that took root in Fazio’s computers.11
The Citadel Trojan horse was nothing novel—it was a variant of a well-
known malware package called ZeuS and is readily detectable by any major
enterprise anti-virus software. But Fazio lacked the massive security
infrastructure that Target had, allowing the malware to remain undetected
on the Fazio computers. Through the Trojan horse, the hackers obtained
Fazio’s log-in credentials for Target’s system.
With access to Target, the hackers unleashed a different malware
program, one they bought on the black market for just a few thousand
dollars.12 Experts such as McAfee director Jim Walker characterized the
malware as “absolutely unsophisticated and uninteresting.”13
At first, the malware went undetected, and it began compiling millions
of records during peak business hours. This data was being readied to be
transferred to the hackers’ location in Eastern Europe. But very soon,
FireEye flagged the malware and issued an alert. Target’s security team in
Bangalore noted the alert and notified the security center in Minneapolis.
But the red light was ignored.
FireEye flagged as many as five different versions of the malware. The
alerts even provided the addresses for the “staging ground” servers, and a
gaffe by the hackers meant that the malware code contained usernames and
passwords for these servers, meaning Target security could have logged on
and seen the stolen data for themselves.14 Unfortunately, the alerts all went
unheeded. Furthermore, given that several alerts were issued before any
data were actually removed from the Target systems, FireEye’s automated
malware deletion feature could have ended the assault without the need for
any human action. However, the Target security team had turned that
feature off, preferring a final manual overview of security decisions.15
With FireEye’s red lights blinking furiously, the hackers began moving
the stolen data on December 2, 2013. The malware continued to exfiltrate
data freely for almost two weeks. Law enforcement officials from the
Department of Justice contacted Target about the breach on December 12,
armed not only with reports of fraudulent credit card charges, but also
actual stolen data recovered from the dump servers, which the hackers had
neglected to wipe.16
The aftermath of the breach caused tremendous financial damage to
Target. It remains unknown what the precise cost of the breach was, but an
estimate in Target’s annual report of March 2016 put the figure at $291
million.17 The company’s reputation was harmed. The CIO resigned. For
customers, there was increased risk of future fraud. Daily spending and
withdrawal limits had to be placed on many affected accounts, and new
credit cards had to be issued, causing consumers significant time loss while
updating their card information everywhere.18
The breach went down in the annals of data breach history—one for the
record books. But it would soon be overshadowed by even bigger breaches.
THE SYSTEM IS DOWN
On paper, the hackers never should have been able to breach Target. The
hackers used cheap methods, such as readily detectable malware that wasn’t
state-of-the-art. They were quite sloppy and made careless mistakes. Target
had much better technological tools and a large and sophisticated team. It
conducted phishing tests and employed forensic investigators. The hackers
were grossly outspent and outnumbered. Yet Target was still felled.
At first glance, it seems that Target’s Achilles’ heel was one employee at
one of its third-party vendors. Most large companies have hundreds of
third-party vendors. This person made just one wrong click of the mouse,
and that was all the hackers needed. Had that one person not clicked, then a
data breach leading to more than half-a-billion dollars might not have
occurred. That’s one very expensive mouse click!
However, a prolonged look reveals a host of systemic vulnerabilities.
Although on a checklist Target looked healthy, it lost because one key factor
wasn’t accounted for—human behavior. Spending millions of dollars and
installing high-tech software still couldn’t prevent the humans from their
fateful blunders.
It doesn’t necessarily take technical wizardry or great skill to be a highly
successful criminal on the Internet. Technologies and data ecosystems are
so fragile and flawed that it is far too easy for hackers to break in. The black
market is overflowing with cybercrime start-up kits.19 Just download the
tools and it’s off to the races. Because crime committed using the Internet is
rarely tracked down and enforced, in most cases, the fraudsters get away
with it.
WE HAVE MUCH TO LOSE
The stakes for data security are enormous. Data breaches, by which we
mean the unauthorized exposure, disclosure, or loss of personal
information, are not only more numerous; they are more damaging. Every
year, millions of people are victimized by identity theft. Their personal data
is used by fraudsters to impersonate them. Victims suffer because their
credit files become polluted with delinquent bills. Creditors go after victims
for the unpaid bills, and victims struggle to prove that the bills weren’t