October 02, 2004

Identity theft - buy a Mac, download Firefox

In the "war on phishing" which has yet to be declared, there is little good news. It continues to increase, identity theft is swamping the police departments, and obscure efforts by the RIAA to assert that CD pirating is now linked to financing of terrorism grab the headlines [1]. Here's a good article on the victims, and the woe that befalls the common man of the net, while waiting for something to be done about it [2].

Meantime, what to do? Phishing won't stop until the browser manufacturers - Microsoft, Mozilla, Konqueror, Opera - accept that it's an attack on the browser. The flood of viruses on Microsoft's installed base won't change any time soon, especiallly underscored by the SP2 message: Microsoft has shown there is no easy patch for a fundamentally broken system.

Don't hold your breath, it will take years. In the meantime, the only thing I can think of for the embattled ordinary user is this: buy a Mac and download Firefox. That won't stop the phishing, but at least they are sufficiently inured against viruses that you won't have to worry about that threat.

[1] http://go.theregister.com/news/2004/09/28/terrorist_email_scams/
[2] http://www.theregister.co.uk/2004/09/24/identity_snatchers/

Invasion of the identity snatchers
By Kelly Martin, SecurityFocus (kel at securityfocus.com)
Published Friday 24th September 2004 11:32 GMT

Last year I was the victim of identity theft, a sobering reality in today's world. An unscrupulous criminal managed to social engineer his way past the formidable security checks and balances provided by my credit card company, my bank, and one of my investment accounts. He methodically researched my background and personal information until he could successfully impersonate me, and then subsequently set forth to change the mailing addresses of my most important financial statements.

It was a harrowing experience, and one worth explaining in the context of the online world. Numerous visits to the local police and the Canadian RCMP revealed some rather surprising things: identity theft is already so common that there are entire units within law enforcement that deal with this issue every day. They have toll-free numbers, websites and documents that clearly define their incident response procedures. But the reality is, law enforcement will respond to these issues just as you might expect: with phone calls, in-person interviews, and some traditional detective work. It's still very much an analog world around us.

The other thing that became crystal clear during the process of regaining my own identity is this: for as capable as they may be, law enforcement is woefully ill-equipped to track down identity theft that starts online. As a security professional with a healthy dose of paranoia, I was confident that my online identity had not been compromised - a more traditional approach had been used. But with the sophistication of today's viruses, millions of others cannot say the same thing.

While not all identity theft starts online, the fact is that online identity theft is now incredibly easy to do. The same methodical, traditional approach that was used to steal my identity by placing phone calls is being sped up, improved upon, and made ever more lethal by first attacking the victim online. Your banking and credit card information can come later.

We all know how commonplace these technologies already are: keyloggers, Trojans with remote-control capabilities and even webcam control, and backdoors that give access to all your files. There are millions of these installed on infected machines all over the world, lurking in the shadows.

Ever do your taxes on your home computer? All it takes is one Social Insurance Number (or Social Security Number in America), plus some really basic personal information, and you're sunk. Every nugget of information can be worth its weight in gold if, for example, that online banking password that was just logged enables someone to change your address and then, a month later, take out a loan in your name.

The rise of phishing scams over the past two years alludes to this growing menace: your personal information, especially your banking and credit card information, has significant value to a criminal. No surprise there.

Working in the security field, many of us know people who are regularly infected with viruses, worms, Trojans. When it gets bad enough, they reformat and reinstall. I can't count the number of times I've heard people tell me that they're not overly concerned, as they believe that the (often, minimal) personal information on their computer is not inherently valuable. They've clearly never had their personal information put to ill use.

As I was reading the new Threat Report from Symantec, which documents historical virus trends, only the biggest numbers jumped out at me. The average time from vulnerability to exploit is now just 5.8 days. Some 40 per cent of Fortune 100 companies had been infected with worms over a period of six months. There were 4,496 new Microsoft Windows viruses discovered in six months, or an average of 24 new viruses every day. Basically, the epidemic is out of control.

With a few exceptions, however, the most popular and most prominent viruses and worms are not the ones that will be used to steal your identity. It's that carefully crafted email, or that feature-rich and bloated Trojan, that will be used in covert attempts.

Perhaps a suitable solution to the epidemic is a rather old one, and one that I employ myself: encryption of all the personal data that is deemed valuable. I'm not talking about your pictures of Aunt Tilly or your music archive - I'm referring to your tax returns, your financial information, your bill payments, etc. This approach still won't avoid the keyloggers or that remote control Trojan that's sitting on your drive, but it does help to avoid new surprises and mistaken clicks.

And to those users out there whom we deal with everyday and who still say there's nothing important on their computer that requires them to care about today's worms, Trojans, viruses, and so on, the day their own information is stolen and used against them is growing ever more near.

Copyright 2004, SecurityFocus logo (http://www.securityfocus.com/)

Posted by iang at October 2, 2004 10:22 AM | TrackBack

Speaking of which, I am very concerned about that myself and that is one reason I still don't have internet or even a computer at home: I trust the network admins at work to be at least marginally competent whereas I know I am not. That will soon have to change, however, and now I am _very_ worried. One thing I tought of and which I have since learnt is used by the NSA is to perform all dangerous activities (like connecting to the internet) in a virtual machine (e.g., using VMWare if on a PC) that has limited privileges and is regularly destroyed (i.e., you keep a copy of the original, never-used file around and you just overwrite the just-used VM instance with it: same as reformatting and reinstalling but it takes only a few seconds, so you can do it as often as you wish: there is no deterrent). The problem is that the VMs and the host still share some hardware (the Ethernet card, for one) and I am worried that it must be easy to get it wrong and get only fictitious security as a result. Because, e.g., NSA uses it I know my scheme is basically sound but I need to get the details right. The fact that recently I wasn't even able to get X running on a Linux instance I installed under VMWare is not encouraging: obviously I simply do not have the skills to get this right. So my question is: would you know per chance know where I can find a howto for implementing this scheme correctly?

An alternative is to have several computers and not to network the one used to connect to the internet but that gets expensive and is rather inconvenient as well.

Posted by: Olivier at October 3, 2004 01:14 PM

Why is the name "McCarthy" running through my head so many a times these days ?

Like a song of old, you hate the tune, but it is too easy to hum along with it.


Posted by: Twan at October 3, 2004 01:15 PM

So, no idea at all?

-- O.L.

Posted by: Olivier at October 9, 2004 04:19 AM

I personally don't know where there is one. Places to check would be the general security groups, and anything to do with OpenBSD.

That sort of scheme is ... if it works ... most likely to be related to the OpenBSD operating system. This doesn't mean that it would only be available in OpenBSD, but if you want to place security before anything else, that's where you'll find the people who talk about it.

When it comes to installing VMWare and running virtual machines, there are limitations to those sorts of systems. The PC architecture is not that clean and easy. Things like X dive in and muck with the hardware directly, which means that a virtual machine has a tough job in trapping and delivering that interface. I'm not that surprised that VMWare couldn't easily do the X, and even if it could be got going, I'd be suspicious of it breaking the first time you get some delta (new app, new hardware, new OS...).

Posted by: Iang at October 9, 2004 04:48 AM