Sovereign Bank and PenFed warn customers after keyloggers are found on laptops

Recent data breaches at two banks underscore what's becoming a gnarly problem for companies that handle sensitive information: When does a hacked PC become a data breach?

Sovereign Bank noticed its problem on Oct. 15, when staffers discovered a computer on their network connecting to an unusual IP address. After investigating, they found a keylogger program on a company laptop. Sovereign isn't releasing many details on the incident, but in December it notified 50 customers nationwide that their data may have been compromised.

Over at Pentagon Federal Credit Union, a bank used by nearly 1 million US service members, a hacked laptop led to a bigger problem. On Dec. 12, the company found that someone had hacked a laptop on its network and used it to access a company database that contained credit card numbers, addresses, Social Security numbers and other sensitive information. A PenFed spokeswoman wouldn't say how many customers were affected, but the company is re-issuing 514 credit cards in New Hampshire alone.

Both incidents underscore how easily a hacked laptop can be used to gain access to sensitive information, an issue that's becoming more of a problem for corporate IT as criminals continue to hit workers with malicious e-mail and links to drive-by-download websites, the two most popular hacking techniques to get malicious software installed on a computer. Criminals set up drive-by-download websites to install malicious software on victims' computers. Typically they leverage known computer flaws to silently install their malicious software.

According to experts, Sovereign's decision to investigate the situation and then notify customers is probably more cautious than most.

The problem is that it's often unclear whether hackers were able to access sensitive data.

In some companies, especially in smaller companies in less-closely regulated industries, IT staff coming across a hacked computer may simply wipe the system and set it up anew from scratch, unaware of any regulatory obligation. It takes a careful forensic audit to even have a chance of figuring out when the malicious software was installed, and whether it was used to access sensitive data. But even if there's a forensic assessment, figuring out whether data has really been accessed can be a matter of guesswork. "There's no right answer," said Alan Cox, a principal research analyst with network monitoring company Netwitness. "It's typically handled on an incident-by-incident basis."

But even as companies struggle to figure out the right response to this type of hacking, the problem is growing. Keyloggers are becoming a common problem in any type of company, including banks, Cox said. "It's one of the top ways that the cybercriminals ... tend to operate," he said. "You can just get so much information."

Usually, it's the lawyers who decide whether to notify customers. And how they make that call can depend on a lot of things: the regulatory environment they're in, the size of the company and the company's own corporate culture.

Often, the decision can be a tough call for the lawyers to make. It costs a lot of money to send out notification letters and provide credit protection after a breach. But if a company's response isn't strong enough, its public image can take a hit. There can be big fines too. Last year, the Office of the Attorney General of the State of Connecticut fined medical insurance provider Health Net $250,000, in part, because the company did not respond properly to a May 2009 data breach.

"There are a lot of breaches that never get to the point of reporting," said Doug Pollack, chief marketing officer at ID Experts, a Portland, Oregon, company that specializes in helping clean up data breaches. "What you actually see in terms of reported breaches is literally only the tip of the iceberg in terms of what's taking place."