Security

This post was updated at 2:16 p.m. PDT with comment from an outside database security software vendor.

Hackers broke into the University of California at Berkeley's health services center computer and potentially stole the personal information of more than 160,000 students, alumni, and others, the university announced Friday.

At particular risk of identity theft are some 97,000 individuals whose Social Security numbers were accessed in the breach, but it's still unclear whether hackers were able to match up those SSNs with individual names, Shelton Waggener, UCB's chief technology officer, said in a press conference Friday afternoon.

(Credit: University of California at Berkeley)

The attackers accessed a public Web site and then bypassed additional secured databases stored on the same server. In addition to SSNs, the databases contained health insurance information and non-treatment medical information, such as immunization records and names of doctors patients had seen. No medical records (i.e. patient diagnoses, treatments, and therapies) were taken, as they are stored in a separate system, emphasized Steve Lustig, associate vice chancellor for health and human services.

"Their ID has not been stolen," he added. "Some data has been stolen."

The server breach began on October 9, 2008, and continued through April 9, when a campus computer administrator doing routine maintenance discovered messages left by the attackers. Logs indicate that the hacks originated from overseas, "primarily in the Asian theater," Waggener said, later specifying traces to China.

While campus police and the FBI were immediately notified of the breach, it wasn't until April 21, Waggener said, that officials learned data had been stolen. Since then, the focus of the investigation has been figuring out what was taken and who is at risk. The hackers' specific techniques are still being determined as part of the ongoing criminal investigation, he said.

From the looks of it, however, one outside database security software vendor, Sentrigo CTO Slavik Markovich, suspects an SQL injection, in which a small malicious script is inserted into a database that feeds information to the Web site. Markovich also questions whether the university has appropriate monitoring tools in place to have not noticed the hack for six months, and why it hosted data with different levels of sensitivity on the same server.

The university started notifying the 160,000 people at risk via e-mail and snail mail on Friday. Victims include an assortment of current and former Berkeley students--as well as their parents or spouses, if linked to insurance coverage--who had University Health Services health care coverage or received services. Also included are 3,400 students of Mills College in Oakland, Calif., which contracts with the university for health services.

The university has warned those affected to put a fraud alert on their credit reporting accounts. It has also set up a Web site and hotline to help the victims.

In 2005, a PC was stolen from a Berkeley graduate admission office that held sensitive data on some 98,000 people, stretching back three decades. And the university has dealt with security viruses and the like, Waggener said. But this was the first such server breach.

With this, Waggener said, Berkeley joins a long list of prestigious institutions suffering from such increasingly sophisticated and malicious attacks. "We're defending against attacks from around the world," he said.

Update at 9:30 a.m. PST: A new chart has been added to the end of the article.

This was originally published in ZDNet's Between the Lines.

Identity theft cases surged in 2008, according to the Federal Trade Commission.

Last year, ID theft was by far the biggest complaint to the FTC, representing 26 percent of total problems reported. The next biggest one--third-party and creditor debt collection scams--represented only 9 percent of complaints.

The FTC's annual Consumer Sentinel Network report (PDF), released Thursday, details that ID theft complaints totaled nearly 314,000 in 2008, up from about 259,000 in 2007 and up substantially from about 31,000 in 2000.

The Consumer Sentinel Network is a secure online database that harvests complaints from law enforcement authorities, as well as other groups such as the Internet Crime Complaint Center and Better Business Bureau.

(Credit: FTC)

Here are the top 10 complaint categories, which often dovetail with the Internet.

(Credit: FTC)

E-mail is clearly the preferred means of propagating fraud. Scam artists are most likely going to nail you via e-mail. Phone scams have fallen from 11 percent to 7 percent from 2007 to 2008. My hunch: as more consumers use wireless as their primary phone, it's harder to track down victims.

(Credit: FTC)

What's also notable: the demographics. Twenty-somethings are most likely to get hit with ID theft.

(Credit: FTC)

Sometime on October 14, a wide array of furniture and electronics were stolen from a commercial storage facility outside Phoenix. The building was used by the Arizona Early Intervention Program, which helps families of disabled children.

Two weeks ago, the state informed the parents of the nearly 40,000 children in the program that their personal information was potentially at risk for ID fraud. According to the Arizona Department of Economic Security (DES), a backup computer hard drive stolen from the facility was password protected. What happened next is where the controversy arises.

The DES and others in the media suggested that parents concerned about protecting their children against ID fraud seek a credit report for each child, and then put a credit freeze on the credit bureau accounts--advice that initially sounded right to me. But sources tell CNET News that such steps are wrong. Jay and Linda Foley, of the Identity Theft Resource Center (ITRC), said ordering a credit report that technically should not exist is one of the worst things you can do.

Making the problem worse
Julie Fergerson, vice president of emerging technologies at Debix, agreed. "If you actually try to order the credit file, there is a certain number of inquires against the Social Security number that the credit bureaus will create, potentially, on accident, a credit file."

Scott Mitic, CEO of TrustedID said, "according to the Federal Trade Commission, as many as 400,000 children may already be victims of identity theft. To make matters worse, the number of complaints has increased by 78 percent over the past several years, making children the fastest growing segment of identity theft victims." He said common warning signs include the receipt of pre-approved credit offers addressed to your child, calls from a collection agency in which the caller asks for your child by name, or notices addressed to your child from government or law enforcement agencies.

Scott Mitic, CEO of TrustedID

(Credit: TrustedID)

Tom Rusin, president and chief executive officer of Affinion's North America operation, said there should be no credit information being stored for minors with the credit bureaus, but they aren't consistent with what age they start to hold a child's information. "For some they hold information for those 18 and older, with one it's 16 and older. Technically speaking, if you are nine, your information should not reside within the credit bureaus at all."

When is too early?
Children today can get a Social Security number assigned within days of birth. That number may be valuable for setting up college saving accounts and obtaining company health benefits, but, in most cases, that Social Security number sits dormant for about 16 years. No loans. No credit cards. No activity. Pat Dane, chief revenue officer at MyPublicInfo, recommends "as soon as the parents give the kid a Social, they ought to start monitoring it."

"It's a squishy area," said Affinion's Rusin. "If they don't have credit files, how can you monitor them?"

So what kind of monitoring is right for a child?

Julie Fergerson, Debix VP of emerging technologies

(Credit: Debix)

Not traditional credit report monitoring, warned ITRC's Jay Foley. He said it's not a good idea to sign up a child for a service for something that does not exist.

Debix's Fergerson told me when ID theft occurs among children, a credit file is often attached to the child's Social Security number with the suspect's name and date of birth, not the child's. "So doing the traditional things like ordering fraud alerts or credit reports, any of those things, will always come back saying there is nothing there."

Mike Prusinski, VP of public affairs at LifeLock, agreed: "A credit freeze cannot be placed if there is nothing to attach it to. After multiple attempts or inquiries (in)to a child's identity, it is possible that a credit file might be created."

"And if there is a credit report file (associated with your child's name), it's not always necessarily identity theft said ITRC's Linda Foley. "It could be that someone mixed up the numbers and instead of a six they put down a five. And sometimes credit files are created because of clerical errors," said Foley. "The key here is to identify it early so we can fix it."

ID monitoring is not credit monitoring
Different from credit monitoring is ID monitoring. MyPublicInfo's Dane explained to me the subtle distinction between credit monitoring and ID monitoring, the difference that has ID fraud experts upset with those spreading misinformation about protecting children. Credit monitoring and ID monitoring are not the same, said Dane, who sent me some Gartner studies showing that credit report monitoring isn't as effective today as ID monitoring when it comes to detecting new account creation, for example. ID monitoring casts a much wider net, looking for activity on a person's Social Security number, not their credit report.

"If someone stole my son's Social," he said, "they could walk into Verizon, T-Mobile and open the easiest form of credit there is." Establishing a utility record is a common way that identity fraud is committed in part because it is harder to identify. Instead of appearing on a credit report, it needs a separate monitoring process, which the Gartner reports say most people do not have. When this so-called "synthetic ID theft" happens to a child, it may occur for years and years before the child needs to establish credit and finds he or she cannot.

"To me (new account creation) is probably one of the more egregious forms of identity theft," ITRC's Linda Foley said.

ITRC's Jay Foley said there's the classic story of a child in foster care. The kid turns 18 and the county ceases supervision. The kid then learns that through a bad parent or other means there's a bad credit report. "Instead of that child going on straight from high school to college, the child's going to end up working low- to pathetic-wage jobs while they clean up this mess in order to qualify for a student loan," he said.

Linda Foley, Founder, ID Theft Resource Center

(Credit: ITRC)

What should you do?
ITRC's Linda Foley said "if you think that your child may be a victim of identity theft, parents need to fire off registered letters to each of the credit bureaus. The letters should include the child's full name, Social Security number, parent (or guardian's) name and address. The letter should ask that a search for a credit file be done of the child's Social Security number since often the name will be different. Additionally you should include photocopies of your driver's license (proof of your identity), a copy of the child's birth certificate showing you as the parent, any guardianship papers if you are not the parent and a copy of the child's Social Security card. Foley said it sounds like a lot, but that's what photocopiers are for.

The credit bureaus want to make sure you are the correct person before releasing information, Foley said. If you are told, "there is no file," that is a good answer and you should stop worrying. Check again when the child is 16 and then again when they are 17 and getting ready to apply for a job or college. "If you are told there is a file, contact one of the non-profits or government agencies that provide victim assistance at no charge," she said. "They will walk you through the steps to clear the records."

LifeLock's Prusinski said for minors 15 and under, his company attempts to set a fraud alert every six months; for children over the age of 16, it is every 90 days, just like adults. "Although we cannot place an actual alert if no credit file exists, we still take the necessary measures to ensure that we are preventing a credit file from being fraudulently created." In addition LifeLock does a credit report audit for minors once a year through the FACT Act, which only requests a credit file. "This action has not created an inquiry because there is nothing with the bureaus that matches that SSN or name." Ideally, parents should then receive the letter that states "a credit file cannot be found." LifeLock also performs a separate Social Security Administration audit for children to see if work history exists.

Debix will also monitor a child's ID and if there's a problem, it'll clean it up. Recently Debix partnered with Javelin research to study the first 500 children who signed up with its service. Of that group, researchers found 5 percent had a pre-existing problem. Debix' Fergerson said that 12 percent were aged 5 and younger, and the average amount of each fraud was about $12,000. She said the company saw one case where a 17-year-old found his Social Security number had been used by a woman for the last two decades, a woman who had $325,000 in debt, a mortgage, and car loan. The 17-year-old boy was a few months away from applying for college. "This case, the woman wasn't a criminal, she legitimately believed the number was hers." Debix straightened out the accounts.

Trusted ID offers similar protection for minors.

Affinion's Rusin said his company is in the process of creating a children's identity protection program.

Tom Rusin, president and CEO of Affinion's North America operation

(Credit: Robert Vamosi / CNET)

Catch it young
Right now parents and guardians cannot put a block on a child's Social Security number saying it "belongs to a minor," but Linda Foley said she's working to make that a federal law by the end of 2009. Affinion's Rusin further suggested that the Social Security Agency also needs to improve its database so that two names don't show up under one SSN.

"The reality is if we catch it when they are young, before they are 16 or 17 years old," Linda Foley said, "it is far easier to take care of than if you were to become a victim of identity theft because we can show that anyone under the age of 18 who is still a minor, not emancipated, cannot be held legally responsible for any contract." Knowing early on makes it easier for parents to repair the situation, she said.

There have been 516 large security breaches this year to date, according to the ID Theft Resource Center, with 30,382,786 consumers potentially exposed.

That number is significant, because it's greater than the record number of breaches reported in 2007, said the ITRC, which released its report (PDF) last week. Part of the rise is the result of increased disclosure to consumers affected by breached organizations, as required by state laws. The ITRC also attributed the increase to its ability to access state attorney general notification lists; these often contain breaches that were not reported via media or other sources.

However, the 30 million customers whose records have been exposed this year may not all be affected by ID fraud; the affected companies may provide credit monitoring services which may stop fraudulent activity before it spirals out of the control. Furthermore, there may be duplicates in the information listed. (One person's information may appear in multiple breaches.)

A February 2007 survey from Javelin Strategy and Research reported around 8 million actual victims of ID Fraud in the U.S., and the amount of fraud had decreased from 2006 to 2007.