TippingPoint had offered $10,000 for each exploit on any of the phones, which included the iPhone and the BlackBerry, as well as phones running the Windows Mobile, Symbian and Android operating systems.
With the mobile devices so limited on memory and processing power, a lot of [researchers’] main exploit techniques are not able to work,‘
The problem being that phones are going to get more powerful rather than less so there’s still work to be done on mobile security. They’re going to get memory and processing rivalling some recent generation desktop and laptop machines. Two years ago, @dressjunkie’s main computer was a 1 GHz Celeron desktop with 256 MB RAM. Her current iPhone is a 667MHz processor running at 412MHz with 128 MB RAM. The next generation iPhone may well run faster (considering the iPod Touch 2nd Generation runs at 533 MHz!).
I guess we can look forward to great exploits in the future. iPhone is pretty much untested here – the only mobile OS to make it to UK GOV CESG testing and survive is RIM’s Blackberry. iPhone hasn’t been tested by CESG and Windows Mobile has been tested and rejected five times.
The article link is broken, which is unfortunate because I’d love to understand the rationale behind “main exploit techniques are not able to work”, which I’d take at face value as BS. The real reason nobody targets these devices is that they are relatively low yield, a little like why OS X enjoyed ‘good’ security for much of its first decade.
It’s easy to find arm devices which are actively being targeted, and very successfully. I know a few people working with routers, in particular on the kind of attack that only this week reached the press (the router-based botnet).
The difference here is that the routers usually have a lot of bandwidth attached, a useful commodity for the average hax0r type. Rifling through people’s iPhone contacts isn’t so appealing.
Sorry, that should have read “exist in very large numbers and have a lot of bandwidth attached.”. The numbers bit is probably a lot more important than the bandwidth.
The link works for me (I just tested it again)- but does deliver an advert first.
It’s not about whether people target them – there are people working on them – but one of the issues specifically mentioned was the lack of resources – which can work both ways of course, depending on whether you want DOS or exploit.
I think it will get worse, we’ll see more exploits. What use they are is immaterial.
I think Google Reader munged the URL or I got served a temporary error.. I see what they’re saying now, and I guess it’s fairly accurate. For some platforms it might only take a single modified default phone welcome screen to screw up the memory offsets required for an exploit to work, etc., but there’s already techniques for ‘regular’ machines that mitigate those problems (e.g. heap spraying).
We’re only really beginning to see the start of mass targeted efforts by organized crime using modern approaches to con people (e.g. targeting e-mail phishing scams to particular locales), pretty certain what we’re witnessing there is a fledgling industry. Definitely agree with you on things heating up.
I’d be willing to place some kind of bet that the mobile space will get vastly less secure as batteries get better and bandwidth becomes cheaper/more available. That, or the Russians start selling their 0days to the Nigerians, who’ve taken to using people’s mobile address books for highly personalized 419 scams 😉
Do they even need a 419 as we give them unfettered access to mobile dial plans, premium SMS rates and even our credit cards built into apps (like the iTunes store).
Anthony Hutton of eyeSpyFX outlined the challenges in coding for 15 different types of mobile phone – mostly running the same OS – when I first met him last year – I guess an exploit will have to deal with the same.