Category Archives: Risk Assessment

OwnStar Wi-Fi attack now grabs BMW, Mercedes, and Chrysler cars’ virtual keys

Remember OwnStar? Earlier this month, security researcher and NSA Playset contributor Samy Kamkar demonstrated a Wi-Fi based attack that allowed his device to intercept OnStar credentials from the RemoteLink mobile application—giving an attacker the ability to clone them and use them to track, unlock, and even remote start the vehicle. Kamkar discussed the details of the attack last Friday at DEF CON in Las Vegas, noting that the RemoteLink app on iOS devices had failed to properly check the certificate for a secure connection to OnStar’s server, or—as is more common in mobile apps using HTTPS to access Web services—use a “pinned” certificate hard-coded into the application itself. OnStar quickly resolved the issue with a RemoteLink app update.

But OwnStar has moved on to other targets. Today, Kamkar announced that he had adapted the tool to target applications for BMW Remote, Mercedes-Benz mbrace, and Chrysler’s Uconnect services on Apple iOS devices. All three, he said in an exchange with Ars via Twitter, have the exact same vulnerability as the RemoteLink app did: “no pinned cert or even PKI/[certificate authority] validation. Trivial to attack an unadulterated mobile device.”

The OwnStar device packs all the components required to execute this attack into a portable case that can be placed near a targeted vehicle. Like a virtual bear trap, it can capture the login credentials of a car owner using a mobile app to remotely unlock, lock, or start the vehicle, which can then be loaded onto a copy of the targeted mobile app on the attacker’s own device—giving the attacker the ability to execute all of the functions of the telematics system on the targeted vehicle. And it’s all because of a flaw that is all too common to mobile applications—reliance on a remote server’s certificate being valid, regardless of what network the connection is over.

Read 3 remaining paragraphs | Comments

Android security on the ropes with one-two punch from researchers

Android security woes got worse on Thursday, with two separate reports of code defects that put millions of end users at risk.

The first involves the update Google released last week fixing a flaw that allowed attackers to execute malicious code on an estimated 950 million phones with nothing more than a maliciously crafted text message. Seven days later, security researchers are reporting that the patch, which has been in Google’s possession since April, is so flawed that attackers can exploit the vulnerability anyway.

“The patch is 4 lines of code and was (presumably) reviewed by Google engineers prior to shipping,” Jordan Gruskovnjak and Aaron Portnoy, who are researchers with security firm Exodus Intelligence, wrote in a blog post published Thursday. “The public at large believes the current patch protects them when it in fact does not.”

Read 6 remaining paragraphs | Comments