Posted by Martijn Grooten on Nov 9, 2017
Last week saw yet another successful edition of Mobile Pwn2Own, the contest in which participants are challenged to attack fully patched mobile devices using previously unknown vulnerabilities.
Contests like these, and their desktop equivalents, serve two purposes: device manufacturers have vulnerabilities responsibly disclosed to them, while offensive security researchers are able to show off their skills, and get a nice cash prize as a bonus.
One possible message to be taken away from the contest is that every single device ended up getting “pwned”; another, however, would be that it has become increasingly hard to do so. In the case of the Samsung Galaxy S8, for example, it took the attackers no fewer than eleven vulnerabilities to execute code and maintain persistence on the device.
But both of these messages miss the point regarding the way in which almost all attacks against mobile devices happen: they don’t exploit known or unknown vulnerabilities, rather they exploit the human factor.
In a typical “attack”, the phone’s owner is tricked into installing an app, either because it is supposedly necessary, because it promises great things, or maybe just because it is indistinguishable from the real app – which is what made one million users install a rogue version of WhatsApp.
And this is also why, no matter what Pwn2Own‘s results may suggest, iPhone remains the more secure mobile operating system. Not because of inherent properties of the system itself, but because its strictly controlled environment does a far better job at protecting its users against themselves, even if this protection comes at a cost: earlier this year, Apple decided its Chinese users needed to be “protected” against the use of VPNs.
This does not make iPhones 100% secure though, and we have seen a few cases where an iPhone zero-day was used to compromise the device of a very few targets. But such attacks are rare, and for almost all users attacks like this are a much smaller risk than the chance of they themselves inadvertently opening up their device to an adversary. This is the reason why, for users that are a high-value target, I always recommend an iPhone or a similarly locked-down device (even if I myself am perfectly happy with my Android phone).
And it is also why I believe that using a third-party security app to augment a device’s security is a very sensible thing to do, especially on Android, even while acknowledging the limited powers such apps have by design.
Both bad news (all devices were pwnd) and good news (pwning is increasingly difficult) came from the most recent mobile Pwn2Own competition. But the practical security risks that come with using mobile phones have little to do with vulnerabilities.
At VB2017 in Madrid, industry veteran and ESET Senior Research Fellow David Harley presented a paper on the state of security software testing. Today we publish David’s paper in both HTML and PDF format.
Trickbot, a banking trojan which appeared this year, seems to be a new, more modular, and more extensible malware descendant of the notorious Dyre botnet trojan. At VB2017, Symantec researcher Andrew Brandt presented a walkthrough of a typical…
Today, we publish a short paper in which CERT Société Générale presents FAME, its open source malware analysis framework.
Ebury and Mayhem, two families of Linux server malware, about which VB published papers back in 2014, are still active and have received recent updates.