Equifax Inc. said today an investigation into information stolen in the epic data breach the company disclosed on Sept. 7 revealed that intruders took a file containing 15.2 million UK records. The company says it is now working to inform 693,665 U.K. consumers whose data was stolen in the attack.
Previously, Equifax said the breach impacted approximately 400,000 U.K. residents. But in a statement released Tuesday, Equifax said it would notify 693,665 U.K. consumers by mail that their personal information was jeopardized in the breach. This includes:
-12,086 consumers who had an email address associated with their Equifax.co.uk account in 2014 accessed.
-14,961 consumers who had portions of their Equifax.co.uk membership details — such as username, password, secret questions and answers, as well as partial credit card details — accessed
-29,188 consumers who had their drivers license numbers accessed
-637,430 consumers who had their phone numbers accessed
The numbers include data that Equifax held on U.K. consumers as far back as 2011, the company said. Equifax did not say whether any of the above-mentioned data was encrypted.
Meanwhile, the U.K.’s National Cyber Security Centre is warning residents to be on their guard against phishing attacks made to look like communications from Equifax about the breach.
“Another risk to UK citizens affected by this data breach is that they could be on the receiving end of more targeted and realistic phishing messages,” the NCSC wrote. “Fraudsters can use the data to make their phishing messages look much more credible, including using real names and statements such as: ‘To show this is not a phishing email, we have included the month of your birth and the last 3 digits of your phone number’. These phishing messages may be unrelated to Equifax and may use more well known brands. It is unlikely that any organisations will ask their customers to reset security information or passwords as a result of the Equifax breach, but this may be a tactic employed by criminals.”
Equifax has been widely criticized for continuously bungling their public response to this still-unfolding data disaster, and today’s update about the extent of the breach in the U.K. was no exception. The Equifax Web site that hosts today’s press release serves “mixed content,” meaning it includes elements that are served over both encrypted and unencrypted pages. The practical effect of this varies depending on which browser you’re using, but some browsers will display a security warning when this happens.
That mixed content error may have something to do with a missing image in the press release. That press release was supposed to include an image that breaks down what exactly was stolen from U.K. residents — as detailed in the bulleted list above — but apparently the graphic was either removed or moved pre- or post-publication. Here’s what the press release looks like in Firefox (Equifax still hasn’t fixed this):
In Internet Explorer:
It’s fairly terrifying when you realize that a company which can’t even issue a press release without managing to omit the most important piece of information in it wields so much power over consumers. Nothing says ‘we care about your security and privacy’ like a message which warns “you got hacked!” and then fails to tell you what that actually means.
I’ve been spending quite a bit of time looking at Equifax’s various Web properties over the past few weeks and I have to say it gets scarier the more I look. First it was the discovery that Equifax’s consumer dispute portal in Argentina was protected by nothing more than the username and password “admin/admin.” It’s worth noting that, as mentioned countless times by Equifax’s former CEO in front of several congressional committees last week, the breach of sensitive data on 145.5 million Americans began with lax security at just such a dispute portal (the company declined to say which).
Earlier this week I pointed out that the company’s TALX Web site made it trivial to find the salary history of large chunk of the American population, armed with nothing more than someone’s date of birth and Social Security number (both data points, by the way, that were stolen on 145.5 million Americans, thanks to Equifax). The company responded by taking the site offline a few hours after that story ran on Sunday. That site is still “under maintenance,” according to Equifax.
While Equifax has stressed that it will offer free credit monitoring services to victims of its own breach, it is still using the entire incident to drive traffic to areas of its consumer business that make the company oodles of money, such as “FREE* credit report & score” services for only £14.95 per month. It’s impossible to understand how Equifax could fail to notice the atrocious optics here, unless of course it really doesn’t care.
By the way, if you’re somehow just tuning in to news about this breach, don’t sweat it: Here’s a Q&A that explains what’s at stake and what you should do.
In May, KrebsOnSecurity broke a story about lax security at a payroll division of big-three credit bureau Equifax that let identity thieves access personal and financial data on an unknown number of Americans. Incredibly, this same division makes it simple to access detailed salary and employment history on a large portion of Americans using little more than someone’s Social Security number and date of birth — both data elements that were stolen in the recent breach at Equifax.
At issue is a service provided by Equifax’s TALX division called The Work Number. The service is designed to provide automated employment and income verification for prospective employers, and tens of thousands of companies report employee salary data to it. The Work Number also allows anyone whose employer uses the service to provide proof of their income when purchasing a home or applying for a loan.
The homepage for this Equifax service wants to assure visitors that “Your personal information is protected.”
“With your consent your personal data can be retrieved only by credentialed verifiers,” Equifax assures us, referring mainly to banks and other entities that request salary data for purposes of setting credit limits.
Sadly, this isn’t anywhere near true because most employers who contribute data to The Work Number — including Fortune 100 firms, government agencies and universities — rely on horribly weak authentication for access to the information.
To find out how easy it is to view your detailed salary history, you’ll need your employer’s name or employer code. Helpfully, this page lets you look that up quite easily (although if you opt to list employers alphabetically by the fist letter of the company name, there are so many entries for each letter that I found Equifax’s database simply crashes half the time instead of rendering the entire list).
What’s needed to access your salary and employment history? Go here, and enter the employer name or employer code. After that, it asks for a “user ID.” This might sound like privileged information, but in most cases this is just the employees’s Social Security number (or a portion of it).
At the next step, the site asks visitors to “enter your PIN,” short for Personal Identification Number. However, in the vast majority of cases this appears to be little more than someone’s eight-digit date of birth. The formats differ by employer, but it’s usually either yyyy/mm/dd or mm/dd/yyyy, without the slashes.
Successful validation to the system produces two sets of data: An employee’s salary and employment history going back at least a decade, and a report listing all of the entities (ostensibly, the aforementioned “credentialed verifiers”) that have previously requested and viewed this information.
Once you’re successfully “authenticated,” the system asks you to change your PIN to something more secret than your birthday. When the default PIN is changed, The Work Number prompts users to select a series of six challenge/response questions, which Equifax claims will “improve the security of your data and create an extra layer of protection on your account.”
Unfortunately, consumers whose employee history is stored by this service effectively have no privacy or security unless they possess both the awareness that this service exists and the forethought to access their account online before identity thieves or others do it first.
The Work Number does allow employers to opt for TALX’s “enhanced authentication” feature, wherein after logging in with your employer ID and PIN (often the last four digits of an SSN plus the birth year), the system is designed to require the requester to respond to an email at a work address or a phone call to a work number to validate the login.
However, I did not find this to be the case in several instances involving readers whose employers supposedly used this enhanced authentication method. In cases where corporate human resources departments fail to populate employee email addresses and phone numbers, the system defaults to asking visitors to enter any email address and phone number to complete the validation. This is detailed here (PDF), wherein The Work Number states “if you do not have the required phone and e-mail information on file, you will be prompted to update/add your phone numbers/email addresses.”
Many readers probably consider their current and former salaries to be very private information, but as we can see this data is easily available on a broad spectrum of the working population in America today. The information needed to obtain it has been widely compromised in thousands of data breaches over the past few years, and the SSN and DOB on most Americans is for sale in a variety of places online. In short, if you can get these details from Equifax’s online service, so can anyone else.
Fortunately, you can reduce the likelihood that an acquaintance, co-worker, stalker or anyone else can do this by claiming your own account, changing the PIN and selecting a half-dozen security questions and answers. As always, it’s best not to answer these questions truthfully, but to input answers that only you will know and that can’t be found using social networking sites or other public data sources.
I could see this service potentially helping to create a toxic workplace environment because it offers a relatively simple method for employees to glean data about the salaries of their co-workers and bosses. While some people believe that companies should be more transparent about employee salaries, this data in the wrong hands very often generates a great deal of resentment and hostility among co-workers.
Employers who use The Work Number should strongly consider changing as many defaults as possible, and truly implementing the service’s enhanced authentication features.
October is National Cybersecurity Awareness Month, and as such KrebsOnSecurity will continue pointing readers to similar services that let anyone access your personal data armed with little more than static identifiers about you that should no longer be considered private. Although some readers may take issue with my pointing these out — reasoning that I’m only making it easier for bad people to do bad things — it’s important to understand that knowledge is half the battle: Planting your flag before someone else does is usually the only way to keep others from abusing such services to expose your personal information.
–USPS ‘Informed Delivery’ is Stalker’s Dream
–Student Aid Tool Held Key for Tax Fraudsters
–Sign Up at IRS.gov Before Crooks Do It For You
–Crooks Hijack Retirement Funds via SSA Portal
–Social Security Administration Now Requires Two-Factor Authentication
–SSA: Ixnay on txt msg reqmnt 4 e-acct, sry
Maybe you’ve been feeling left out because you weren’t among the lucky few hundred million or billion who had their personal information stolen in either the Equifax or Yahoo! breaches. Well buck up, camper: Both companies took steps to make you feel better today.
Yahoo! announced that, our bad!: It wasn’t just one billion users who had their account information filched in its record-breaking 2013 data breach. It was more like three billion (read: all) users. Meanwhile, big three credit bureau Equifax added 2.5 million more victims to its roster of 143 million Americans who had their Social Security numbers and other personal data filched in a breach earlier this year. At the same time, Equifax’s erstwhile CEO informed Congress that the breach was the result of even more bone-headed security than was first disclosed.
To those still feeling left out by either company after this spate of bad news, I have only one thing to say (although I feel a bit like a broken record in repeating this): Assume you’re compromised, and take steps accordingly.
If readers are detecting a bit of sarcasm and cynicism in my tone here, it may be that I’m still wishing I’d done almost anything else today besides watching three hours worth of testimony from former Equifax CEO Richard Smith before lawmakers on a panel of the House Energy & Commerce Committee.
While he is no longer the boss of Equifax, Smith gamely agreed to submit to several day’s worth of grilling from legislators in both houses of Congress this week. It was clear from the questions that lawmakers didn’t ask in Round One, however, that Smith was far more prepared for the first batch of questioning than they were, and that the entire ordeal would amount to only a gentle braising.
Nevertheless, Smith managed to paint a even more dismal picture than was already known about the company’s efforts to secure the very data that makes up the core of its business. Helpfully, Smith clarified early on in the hearing that the company’s customers are in fact banks and other businesses — not consumers.
Smith told lawmakers that the breach stemmed from a combination of technological error and a human error, casting the breach as the kind of failure that could have happened to anyone. In reality, the company waited 4.5 months (until after it discovered the breach in late July 2017) to fix a dangerous security flaw that it should have known was being exploited on Day One (~March 6 or 7, 2017).
“The human error involved the failure to apply a software patch to a dispute portal in March 2017,” Smith said. He declined to explain (and lawmakers inexplicably failed to ask) how 145.5 million Americans — nearly 60 percent of the adult population of the United States — could have had their information tied up in a dispute portal at Equifax. “The technological error involved a scanner which failed to detect a vulnerability on that particular portal.”
As noted in this Wired.com story, Smith admitted that the data which was compromised in the breach was not encrypted:
When asked by representative Adam Kinzinger of Illinois about what data Equifax encrypts in its systems, Smith admitted that the data compromised in the customer-dispute portal was stored in plaintext and would have been easily readable by attackers. “We use many techniques to protect data—encryption, tokenization, masking, encryption in motion, encrypting at rest,” Smith said. “To be very specific, this data was not encrypted at rest.”
It’s unclear exactly what of the pilfered data resided in the portal versus other parts of Equifax’s system, but it turns out that also didn’t matter much, given Equifax’s attitude toward encryption overall. “OK, so this wasn’t [encrypted], but your core is?” Kinzinger asked. “Some, not all,” Smith replied. “There are varying levels of security techniques that the team deploys in different environments around the business.”
Smith also sought to justify the company’s historically poor breach response after it publicly disclosed the breach on Sept. 7 — roughly 40 days after it says Equifax’s security team first became aware of the incident (on July 29). As many readers here are well familiar, KrebsOnSecurity likened that breach response to a dumpster fire — noting that it was perhaps the most haphazard and ill-conceived of any major data breach disclosure in history.
Smith artfully dodged questions of why the company waited so long to notify the public, and about the perception that Equifax sought to profit off of its own data breach. One lawmaker noted that Smith gave two public speeches in the second and third weeks of August in which he was quoted as saying that fraud was a “a huge opportunity for Equifax,” and that it was a “massive, growing business” for the company.
Smith interjected that he had “no indication” that consumer data was compromised at the time of the Aug. 11 speech. As for the Aug. 17 address, he said “we did not know how much data was compromised, what data was compromised.”
Follow-up questions from lawmakers on the panel revealed that Smith didn’t ask for a briefing about what was then allegedly only classified internally as “suspicious activity” until August 15, almost two weeks after the company hired outside cybersecurity experts to examine the issue.
Smith also maneuvered around questions about why Equifax chose to disclose the breach on the very day that Hurricane Irma was dominating front-page news with an imminent landfall on the eastern seaboard of the United States.
However, Smith did blame Irma in explaining why the company’s phone systems were simply unable to handle the call volume from U.S. consumers concerned about the Category Five data breach, saying that Irma took down two of Equifax’s largest call centers days after the breach disclosure. He said the company handled over 420 million consumer visits to the portal designed to help people figure out whether they were victimized in the breach, underscoring how so many American adults were forced to revisit the site again and again because it failed to give people consistent answers about whether they were affected.
Just a couple of hours after the House Commerce panel hearing ended, Politico ran a story noting that the Internal Revenue Service opted to award Equifax a $7.25 million no-bid contract to provide identity-proofing and anti-fraud services to the tax bureau. Bear in mind that Equifax’s poor security contributed to an epidemic of tax refund fraud at the IRS in the 2015 and 2016 tax years, when fraudsters took advantage of weak security questions provided to the IRS by Equifax to file and claim phony tax refund requests on behalf of hundreds of thousands of taxpayers.
Don’t forget that tax fraudsters exploited this same lax security at Equifax’s TALX payroll division to steal employee tax records from an as-yet undisclosed number of companies between April 2016 and March 2017.
Finally, much of today’s hearing centered around questions about the difference between a security freeze — a right that was hard-won on a state-by-state level over several years — and the “credit lock” services being pushed instead by Equifax and the big bureaus. Lawmakers on today’s panel seemed content with Smith’s answer that the two things were effectively the same, only that a freeze was more cumbersome and costly, whereas credit locks were free and far more consumer-friendly.
To those still wavering on which is better, I have only to point to reasoning by Christina Tetreault, a staff attorney on the financial services team of Consumers Union — the policy arm of Consumer Reports. Tetreault notes that perhaps the main reason a security freeze is the better option is that its promise to guard your credit accounts is guaranteed by law, whereas a credit lock is simply an agreement between you and the credit monitoring company.
“Having a contractual agreement is not as strong as having protections under law,” Tetreault said. “The contract may be unclear, may include provisions that allow the other party to change it, or include provisions that you may be better off not agreeing to, such as an arbitration agreement.”
What’s more, placing a freeze on your file is exactly what Equifax and the other bureaus do not want you to do, because it prevents them from making money by selling your credit file to banks and others (including ID thieves) who wish to grant new lines of credit in your name. If that’s not the best reason for opting for a freeze, I don’t know what is.
If anyone needs more convincing on this front, check out the testimony given in other committees today by representatives from banking behemoth Wells Fargo, which is under fire signing up tens of thousands of auto loan customers for insurance they did not need and in some cases couldn’t afford. That scandal comes on the heels of another debacle in which Wells Fargo was found to have created more than 3.5 million bank accounts without consumers’ permission between 2009 and 2016.
Mr. Smith is slated to testify before at least three other committees in the House and Senate this week before he’s off the hot seat. On Friday, KrebsOnSecurity published a lengthy list of questions that lawmakers should consider asking the former Equifax CEO. Here’s hoping our elected representatives don’t merely use these additional opportunities for more grandstanding and regurgitating the same questions.
A free new service from the U.S. Postal Service that provides scanned images of incoming mail before it is slated to arrive at its destination address is raising eyebrows among security experts who worry about the service’s potential for misuse by private investigators, identity thieves, stalkers or abusive ex-partners. The USPS says it hopes to have changes in place by early next year that could help blunt some of those concerns.
The service, dubbed “Informed Delivery,” has been available to select addresses in several states since 2014 under a targeted USPS pilot program, but it has since expanded to include many ZIP codes nationwide, according to the Postal Service. U.S. residents can tell if their address is eligible by visiting informeddelivery.usps.com.
According to the USPS, some 6.3 million accounts have been created via the service so far. The Postal Service says consumer feedback has been overwhelmingly positive, particularly among residents who travel regularly and wish to keep close tabs on any mail being delivered while they’re on the road.
But a review of the methods used by the USPS to validate new account signups suggests the service is wide open to abuse by a range of parties, mainly because of weak authentication and because it is not easy to opt out of the service.
Signing up requires an eligible resident to create a free user account at USPS.com, which asks for the resident’s name, address and an email address. The final step in validating residents involves answering four so-called “knowledge-based authentication” or KBA questions. KrebsOnSecurity has relentlessly assailed KBA as an unreliable authentication method because so many answers to the multiple-guess questions are available on sites like Spokeo and Zillow, or via social networking profiles.
Once signed up, a resident can view scanned images of the front of each piece of incoming mail in advance of its arrival. Unfortunately, because of the weak KBA questions (provided by recently-breached big-three credit bureau Equifax, no less) stalkers, jilted ex-partners, and private investigators also can see who you’re communicating with via the Postal mail.
Perhaps this wouldn’t be such a big deal if the USPS notified residents by snail mail when someone signs up for the service at their address, but it doesn’t.
Peter Swire, a privacy and security expert at Georgia Tech and a senior counsel at the law firm of Alston & Bird, said strong authentication relies on information collected from multiple channels — such as something you know (a password) and something you have (a mobile phone). In this case, however, the USPS has opted not to leverage a channel that it uniquely controls, namely the U.S. Mail system.
“The whole service is based on a channel they control, and they should use that channel to verify people,” Swire said. “That increases user trust that it’s a good service. Multi-channel authentication is becoming the industry norm, and the U.S. Postal Service should catch up to that.”
I also wanted to know whether there was any way for households to opt out of having scanned images of their mail sent as part of this offering. The USPS replied that consumers may contact the Informed Delivery help desk to request that the service not be presented to anyone in their household. “Each request is individually reviewed and assessed by members of the Postal Service Informed Delivery, Privacy and Legal teams,” the Postal Service replied.
There does not appear to be any limit on the number of people who can sign up for the service at any one address, except that one needs to know the names and KBA question answers for a valid resident of that address.
“Informed Delivery may be accessed by any adult member of a household,” the USPS wrote in response to questions. “Each member of the household must be able to complete the identity proofing process implemented by the Postal Service.”
The Postal Service said it is not possible for an address occupant to receive emailed, scanned images of incoming mail at more than one email address. In other words, if you wish to prevent others from signing up in your name or in the name of any other adults at the address, the surest way to do that may be to register your own account and then urge all other adult residents at the address to create their own accounts.
A highly positive story about Informed Delivery published by NBC in April 2017 suggests another use for the service: Reducing mail theft. However, without stronger authentication, this service could let local ID thieves determine with pinpoint accuracy exactly when mail worth stealing is set to arrive.
The USPS says businesses are not currently eligible to sign up as recipients of Informed Delivery. However, people running businesses out of their home could also be the target of competitors hoping to steal away customers, or to pose as partner firms in demanding payment for outstanding invoices.
Informed Delivery seems like a useful service for those residents who wish to take advantage of it. But lacking stronger consumer validation the service seems ripe for abuse. The USPS should use its own unique communications channel (snail mail) to alert Americans when their physical address has been signed up for this service.
Bob Dixon, the executive program director for Informed Delivery, said the Postal Service is working on an approach that it hopes to make available to the public in January 2018 which would allow USPS to send written notification to addresses when someone at that residence signs up for Informed Delivery.
Dixon said that capability will build on technology already in place to notify Americans via mail when a change of address is requested. Currently, the USPS allows address changes via the USPS Web site or in-person at any one of more than 3,000 post offices nationwide. When a request is processed, the USPS sends a confirmation letter to both the old address and the new address.
If someone already signed up for Informed Delivery later posts a change of address request, the USPS does not automatically transfer the Informed Delivery service to the new address: Rather, it sends a mailer with a special code tied to the new address and to the username that requested the change. To resume Informed Delivery at the new address, that code needs to be entered online using the account that requested the address change.
“Part of coming up with a mail-based verification system will also let us do some additional notification that, candidly, we just haven’t built yet,” Dixon said. “It is our intent to have this ready by January 2018, and it is one of our higher priorities to get it done by then.”