iPhone Payment Peril: Mobile Mayhem Omen?
Written by Frank Hayes And Erik ShermanThe iPhone retains everything typed into it through its onscreen keyboard, including payment-card data, for as long as a year. And that penchant for holding onto payment-card data is only the latest in a long line of mobile data catastrophes that are slowly materializing as mobile deployments start in earnest.
Many apps are simply sloppy about the security of sensitive data. Last week (July 27), Citigroup admitted its iPhone mobile banking app stored account numbers and passcodes on the phone. We’re just beginning to understand how little we know about mobile phones and how much more data they retain than we expect.
PCI guidelines and a whole slew of privacy laws are based on the assumption that a retailer might do something bad to expose payment-card data to a thief. A retailer’s logical response in a case like this: “I didn’t do it. The phone’s operating system did.” But that defense might not hold up if the retailer was aware of the problem and did nothing to avoid it.
Further complicating the situation is the fact that there are ways to keep sensitive information out of the keyboard cache. Apple, however, is likely to bounce any app from its iTunes Store that uses such a workaround.
And in the world of insecure apps, Citi isn’t alone. Mobile security vendor Lookout said last week it has examined 300,000 apps and found many that store personal information. For example, 14 percent of free iPhone apps and 8 percent of free Android apps look into user contact records, and about one-third of free apps on both phones track a user’s location.
That cavalier approach to data retention might not be such a concern, except for the increasing ability of thieves to break into smartphones. One demonstration at last week’s Black Hat conference showed how, if an iPhone user did nothing more than visit an infected Web site, a malware payload could be inserted into the phone. (The demonstration’s Web site, jailbreakme.com, just used iPhone security flaws to perform a “jailbreak” to allow the phone to use apps not authorized by Apple. But any malware might be inserted in the same way.)
There are things retailers can do: Encrypt sensitive data that comes into the app; overwrite everything an app temporarily stores; and even create their own secure ways of keying in payment-card data.
Such options are crucial, because too much sensitive data combined with malware that can easily climb into the iPhone and rummage around in user data means almost anything on an iPhone could be stolen by thieves. And for retailers that ask customers for payment-card data, that’s a very troublesome combination.
It gets worse. Chains are tripping over themselves to make sure their mobile applications do not annoy the privacy police. Best Buy’s experiment with Shopkick, for example, trumpeted the fact that its mobile app only receives data from a broadcast signal, rather than sending it, all to play up privacy protections. What if the phone itself trumps all of those efforts and retains lots of private data–or even all of it–on its own?
August 5th, 2010 at 1:56 pm
This reminds me of the bank that had spent a considerable sum installing a state of the art vault. They got burglarized and thieves made up with the vault’s contents. How is that possible? The vault’s door was left open that day…You are only secure as your weakest link. Both the BlackBerry platform and the Android OS (java virtual machine running on linux, dedicated processes) are both considerably more secure than the iPhone. iPhone: pretty yes, secure, no.