Lest you think Facebook is the only security punching bag getting a beating lately, two significant flaws in the Android application platform have been revealed by overseas security teams. Without going into the gory details, each team has found a different way to create a trojanized APK (the file format in which Android apps are delivered) that is indistinguishable from the original. This would allow an app to appear and function normally, but also execute functions like transmitting your passwords, texts, emails on the sly. Google has already put together a fix and distributed a patch to OEM manufacturers, and supposedly they are able to detect this sort of exploit on the Google Play Store.
You need to worry if you “sideload” apps on your Android phone, which is to say you get apps from sources other than Google Play. Keep in mind, even Amazon’s App store counts as a sideloading source, and as of the moment, they aren’t scanning for this vulnerability.
What this means for you:
Even though Google has issued a fix for this particular vulnerability, they can’t force the update upon the millions of Android phones out there affected by this weakness, as that task lies with the phone manufacturers and the carriers. With the exception of avid power-users, most Android users are unaware that their Android OS may be months or years out of date, primarily because cellular carriers insist on selling phones that use a modified version of the OS that does not automatically get updated when Google updates the core version of Android. On top of this, the carriers are notoriously slow in issuing updates. If you are wondering what folks are talking about when they are discussing “Gingerbread”, “Honeycomb”, “Ice Cream Sandwich” and “Jelly Bean”, they are referring to the various versions of Android OS, where Jelly Bean is the latest. Supposedly this exploit exists as far back as “Donut” (ver 1.6).
Even worse, certain older models of Android phones may never get updated, as the carrier has essentially abandoned firmware updates for phones that are “retired” from active support. Users of these phones have essentially two options: root, unlock and update the phone with a custom version of the Android OS developed by the open source community, or buy a new phone. The former option is definitely not for technically-disinclined. Given the gravity of the vulnerability, the carriers may issue patches for the majority of its phones, but I wouldn’t hold your breath.
Until you are able to verify your Android smartphone is running a version of the OS that fixes this vulnerability, don’t sideload applications. If you want to be extra safe, avoid using smartphone apps that transmit sensitive information like banking passwords, pins and other sensitive personal information. As I’ve reiterated before, exercise caution before convenience, especially when it comes to protecting yourself.
Depending on where you live or work, you’ve probably experienced problems with cellphone coverage for one or more carriers, usually due to your geographical (lack of) proximity to a cell tower, or courtesy of construction materials like concrete, lead and steel in between you and your signal. Thanks to the advent of widely available broadband, cellular providers have been able to build small devices called femtocells that can be connected to your internet connection and will significantly improve cellular signal for a specific carrier in a limited range.
While seen as a godsend for the cell-strength deficient, we also now have to regard them as a security risk, thanks to research performed by analysts at iSEC Partners who have allegedly hacked a Verizon network extender to allow them to eavesdrop on any phone call, text message or other information transmitted from the phone through the compromised femtocell. The researchers plan to publicize their findings at the upcoming Black Hat Conference in August, but have declined to share details for obvious security reasons.
What this means for you:
Unfortunately, you can’t tell your cellphone what radio signal source to use. It’s designed to look for the strongest signal and use it. The iSEC researchers claim it would be trivial to build a portable and unobtrusive hacked network extender and place it in a strategic location to capture confidential calls. If you are in the business of confidential information, you probably already know not to take sensitive calls where ever you might be overheard, and if you are a well-informed adult, you probably already know that the NSA could eavesdrop on your conversation regardless of what cell tower was handling your call. But now we are talking about a commercially available device that is cheap, portable, and apparently, hackable. As before, consider carefully the medium you choose for the delivery of your sensitive information, and when in doubt, err on the side of caution rather than convenience.
Image courtesy of Stuart Miles / FreeDigitalPhotos.net
I often encourage my clients to be paranoid about security, but never to the point of throwing the baby out with the bathwater, which is exactly what the Economic Development Agency did two years ago when responding to a report that some of its computers were infected with malware. Due to a mixture of clerical error, poor communication and straight-up inexperience (in a government agency? Imagine that!), the top brass at the EDA received a report that stated over a hundred devices on its network were infected. Believing the technology to be unrecoverable, they proceeded to physically destroy all of it, including mice, keyboards, monitors, printers and other devices that couldn’t be infected with malware, rather than risk the spread of infection, to the tune of nearly $3 million.
What this means for you:
If you’ve ever had a really bad malware infection, you sometimes might hear the technician say, “It’s probably best if we nuke this thing from orbit,” referring to a favorite scene from the movie Aliens. Obviously, your computer is going to be just fine, as he’s actually just talking about wiping out the contents of your hard drive and starting with a fresh install of your operating system. Unless he’s a contractor who lists the EDA as a former client, in which case you might want to show him the door and call someone else.
In all seriousness, a situation like this can easily happen if your organization’s leadership has an incomplete understanding of technology and security. In the above case, a little knowledge and a pinch of common sense could have saved the EDA a lot of money and embarrassment. Continue to be paranoid about security, but only “nuke from orbit” when your company is completely overrun by man-eating aliens. A malware infection, or even a serious security breach, can be handled without slaughtering all those helpless keyboards and mice.
Back in January of this year, I wrote about Facebook’s impending Graph Search feature (“Facebook Graph Search Cutting Bait for Phishers“) which was set to greatly improve its existing feeble search engine as well as outrage privacy watchdogs. Based upon the feedback the developers received from the small test group to which it was originally released, Facebook went back to the drawing board, and have now decided that Graph Search is ready for its debut.
Unlike the search engine we all know and use, Facebook’s new search engine will rely heavily on the various layers of data that it has accumulated on it’s millions of users, allowing you to perform searches that list “friends who like trucks and football” or “single women in Los Angeles who like Ethiopian food”. Obviously, the results are heavily dependent upon how much information everyone shares about themselves on Facebook, but Facebook is confident that the results will be eye opening.
What this means for you:
If you haven’t heard me mention it before, there’s no better time than the present to log into your Facebook account and check your privacy settings, even if you don’t use it often, or you haven’t updated your profile since you created the account oh so many years ago. If you haven’t logged into Facebook in the past year, they have made a lot of changes to settings and security that will probably bewilder the savviest of users. I linked a guide written by the EFF on Facebook’s privacy settings here: “Tighten Up Your Facebook Security”, and Facebook is also taking a more proactive approach by warning you when you log in that Graph Search is coming and provides you a link to your privacy settings.
BlackBerry (formerly RIM) has been struggling in the smartphone market, having recently fallen into 4th place behind even Microsoft’s fledgling foray into that space. Despite the recent release and generally positive reviews of their 10-series phones, the mobile device manufacturer ceded their corporate dominance years ago to the crushing flood of iOS and Android devices primarily because of the company’s failure to stay competitive on the software side. In a move that has analysts scratching their head, BlackBerry is now making a play via software with a new platform called “Secure Work Spaces” which aims to allow for peaceful and secure co-existance of personal and corporate data on smartphones, including iOS and Android devices.
What this means for you:
Corporations struggle with allowing their employees to use corporate phones for business, and vice versa, with corporate phones and personal usage, primarily because the risk of security breaches is much higher on the personal side. BlackBerry’s new platform is designed to create a partition that keeps the two work spaces (see what they did there?) separate, giving enterprises complete control over corporate data without the distasteful invasion and control over the personal aspects of devices. There are other companies working on this same concept, and have been in the space longer, but BlackBerry’s reputation (and probably some nostalgic sentiment) may win the heart’s and minds of corporate IT managers. Seeing as BlackBerry has historically been a company that depends on hardware sales for revenue, many think that BlackBerry is either making a desperate or cunning pivot to the software space, knowing that there is little chance they can recover any ground in the mobile device race.
Facebook offers its users the ability to upload your email contact list, presumably so you can discover which of your friends are on Facebook (that you haven’t already befriended). Once you’ve done this, you also have the ability to download those contacts via an archiving tool called DYI (Download Your Information), that delivers this information via a simple HTML file. Unfortunately, an unintended “bug” in DYI exposed a rather distasteful (though expected) Facebook practice called data correlation. Here’s what happened:
Say you uploaded a contact “[email protected]” to Facebook, but that’s all the data you had on Mr. Smith: just his email address. Another Facebook user also knows Mr. Smith, but also happened to have his phone number and mailing address as well. Facebook’s data correlation practices stores all data on John Smith, regardless of who uploaded it, in a single record, creating a comprehensive data profile on Mr. Smith. See where this is going? Before they fixed this bug, when you went to download your contact info via DYI, not only would you get the email address you knew about, you’d also get any other contact information uploaded by other users, even if you didn’t know the other person who uploaded the contact info about John Smith!
According to Facebook, this data correlation is done to make “Friend” recommendations to you based upon everything it knows about an individual, across its entire store of information.
What this means for you:
It’s not clear whether Facebook intends to notify any of the six million individuals who are affected by this bug, and supposedly this has been fixed so that Facebook users only have access to the data they uploaded minus the data correlation ties Facebook makes in its internal database. According to Facebook, this security bug wasn’t exploited intentionally or maliciously, and it wasn’t possible for anyone using the tool to access information about users they didn’t already have some form of contact info on already.
This does highlight a larger privacy issue that probably won’t be resolved anytime soon, but has been ongoing for Facebook ever since it first appeared. Your friends have access to your PII (Personally Identifiable Information) and regardless of your own personal wishes, you have no ability to control whether or not they share that information, on Facebook or any other social networking site. As is always the case, if you are concerned with the visibility of your personal information on the internet, do regular searches on your name via Google to see what comes up in public, and work back towards the source to remove that information if necessary. Unfortunately, the Internet never forgets, and there is no “100% guaranteed erase” button, so its sometimes impossible to completely remove that data from public view.
Security researchers at Skycure have discovered another weakness in smartphone security, and this could impact you despite whatever security measures you’ve taken personally. Most smartphone operating systems, iOS and Android included, offer the ability to “remember” the SSID’s and passwords of Wifi networks you have accessed with your smartphone, and have the ability to automatically connect to that network the next time you are in range. Skycure has alleged that at least one major carrier, if not all of them, are also pre-programming certain SSID’s into phones straight from the factory, ostensibly to provide customers with a convenient connection with carrier-hosted or sponsored Wifi hotspots. For example, AT&T iPhones allegedly are shipping with the “attwifi” SSID preprogrammed into the phone, and will supposedly automatically join that wifi network, presumably in use by AT&T’s retail storefronts, if it comes across it.
Here’s why this is bad: hackers could spoof any SSID that you’ve set your smartphone to remember and autoconnect, and they’ve got a straight shot at your phone. Normally, this wouldn’t be a problem, as this requires guessing what SSIDs are stored on your phone, and then getting close enough to that phone with the spoofed Wifi network. But with the above, it would be trivial to sit in a crowded mall or any high-traffic walkway, scanning for AT&T iPhones, knowing that some, if not all, will autoconnect to a fake “attwifi” SSID without the owner ever being aware that they just got hacked.
What this means for you:
This exploit seems to be fairly new, and though Skycure claims to have seen this happening in the wild, it’s not widespread, yet. The best course of action is to disable the “autoconnect” setting for any wifi network you have used with your mobile device, whether it be smartphone, tablet or laptop. It will mean a few seconds of inconvenience anytime you are out and about and trying to get internet access, but it may mean the difference between keeping your cellphone secure or getting it hacked.
UPDATE: By default, Android phones will store SSIDs and passwords for any wifi network you add to your phone, and will automatically connect to that network whenever it is range. There is NO way to disable the autoconnect functionality built into the native Android settings. However, you can use an app to control automatic connections. I am currently testing this app, which is “free” but ad-supported. I’ve not tested it long enough to give a recommendation, but it does allow you to toggle the autoconnect functionality on or off per hotspot. On iOS devices, the only way to natively disable the “auto-join” feature is to actually connect to one of the pre-defined hotspots, eg. visit a local AT&T store, and then turn “Auto-join” off for that particular network.
In a public event hosted by the Reddit.com, infamous NSA whisteblower Edward Snowden answered questions posted by Reddit users on a variety of topics. Of particular note was his response to a question about whether encrypting emails would be an effective way to keep the NSA (or anyone else, for that matter) out of your business. Snowden’s response was both heartening and depressing at the same time:
Encryption works. Properly implemented strong crypto systems are one of the few things that you can rely on. Unfortunately, endpoint security is so terrifically weak that NSA can frequently find ways around it.
What this means for you:
Imagine you want to send a package that contains some very valuable items to a friend on the other side of the world. You carefully wrap the items and then lock them in a briefcase, which is in turn handcuffed to an armored guard, who is then transported via armored truck to your friend’s house. He makes sure that the package is put into your friend’s hands and verifies that your friend is indeed who he says he is, and he even calls you to let you know that the package has been delivered safely. This is analogous to using email encryption to send an email to a friend.
Unfortunately, your friend’s house has a broken lock on the front door, and he carelessly leaves the valuable items in plain view of a window that is also unlocked. That’s analogous to the weak endpoint security Snowden at the end of his response.
In other words, it doesn’t matter how much security you engage on your end if your recipients don’t engage in the same level of security. To use another real-world analogy: cyber attacks are like water – they will flow into every nook and cranny, looking for a way in. It doesn’t matter if 99% of the surface it is covering is impenetrable. That last 1% provides the hairline crack needed to seep in and destroy everything from the inside.
Image courtesy of Stuart Miles / FreeDigitalPhotos.net
Proving that sometimes our Congress people come by their paychecks honestly, a bi-partisan privacy caucus led by Joe Barton (Rep. TX) sent a list of questions to Google’s CEO Larry Page, asking him point blank about several privacy issues, including whether or not Google would allow the use of facial recognition technology on the device.
Supposedly, Google has maintained from the start that facial recognition would never be implemented without “strong privacy protections in place.” In a Google+ post Friday, they reiterated this position and stated that Google “…won’t be approving any facial recognition Glassware at this time.”
What this means for you:
By default, Android OS-based devices can only install software via Google’s Play store. Software distributed via Play must go through Google’s approval process, much like apps on Apple’s iTunes store, so you can assume that Google will be true to their word and prevent distribution of facial recognition apps simply by not approving them. However, unlike iPhones, many versions of Android allow “sideloading” of apps with a simple settings change. Sideloading in the Android ecosystem is well established – Amazon.com has an app store that requires sideloading to be enabled, and instructions for enabling this capability are easily found on their website and many, many others.
Bottom line: this is yet another Pandora’s box that won’t be closed. Facial recognition is a reality, and portable, undetectable devices capable of performing this function are only a step away from today’s consumer technology. Technology (and scientific progress in general) advances despite legal or cultural ramifications. One could argue that society only advances in light of controversial technologies like Google Glass. We are only beginning to glimpse the potential of an always connected and much less private world. Google Glass is only one step in a long, uphill climb.
The upcomign Black Hat security conference features a topic that may give traveling iPhone users second thoughts about using a public charging station to juice up their phones. Three security researchers from Georgia Institute of Technology have built a prototype device that can hack an iPhone through the dock connector merely by being plugged in. Supposedly this hack can be accomplished on the latest iOS update, and does not require any interaction from the user, nor does it rely on the device being jailbroken.
What this means for you:
I’ve always viewed public charging stations as being rather sketchy to begin with, especially the ones that charge you for the service and offer “highspeed charging” which could easily fry your phone’s battery if not the device itself. I’d rather spend a few extra minutes locating a regular wall outlet and using my own equipment. Supposedly the prototype that will be demonstrated at the upcoming conference is too big to fit into a standard Apple-branded iPhone charger, but the designers of the device inferred that stealthier versions wouldn’t be hard to produce at all.
Most modern smartphones combine data and power in the same port (Android phones and most tablets also feature this same convenience) so it may not be just iPhones that will be vulnerable to this method of attack. For now, make sure you use chargers you know are safe regardless of what type of mobile device you use, and avoid public charging stations. This particular cow is well on its way out of that barn.