5:25 PM -- Somebody should probably sort out whether online trading is secure.
I've had an Internet stock-trading service account for, if I'm not mistaken, over ten years now. I suppose I've always wondered what the possibilities for attacking such services might be. Even as I moved along the technology trail and wound up in my current gig as CSI director, where I'm basically paid to be cynical about the security of just about everything, I've held what I guess could be called a naïve faith -- you know, because the stakes are so high and the industry is so regulated -- that a tremendous amount of attention must have been paid to locking down communications between the client software running (in a Web browser) on my desktop and the monster data center off somewhere in The Land of Big Money.
I have done no analysis of the particular online brokerages that I use, though I confess I'm a little tempted to turn a packet sniffer loose on my own connection. But I did learn the other day that there are reasons to fear that the protocols used for online training aren't really so very well buttoned up. Here's an excerpt from the salient paragraph in Kelly Jackson Higgins' recent piece on this ('Hacking Capitalism'):
The application-layer FIX (financial information exchange) protocol is used by financial services firms, stock exchanges, and investment banks for automated financial trading. But apps written to the protocol can be vulnerable to denial-of-service, session hijacking, and man-in-the middle attacks over the Internet... says David Goldsmith, CEO of Matasano Security, who will present the firm's new research on FIX at the upcoming Black Hat USA briefings later this month.
Overall, it's just another one of those "are you freaking kidding me?" moments that one gradually grows accustomed to in computer security. Session hijacking? Here's a protocol that is used, in one capacity or another, by every major online financial service you can think of (have a look at the list of FIX adopters), yet it has no strong internal authentication method and relies on SSL for protection when it crosses the Internet.
How bad is this really? Dunno. After all, you wouldn't expect every protocol to reinvent an encrypted tunnel for itself. But for something like stock transactions, I'd sure like an extra level of ongoing authentication to be going on between client and service, some tight ongoing control that makes man-in-the-middle attacks exceedingly hard.
I'd like it -- but that doesn't mean it's there. And it looks as if Goldsmith has some dirt, too. I'm sure looking forward to the Black Hat session, where I plan to ask lots of questions.
Robert Richardson has online trading accounts, but agrees with Benoît Mandelbrot that individual stocks are much more volatile than people generally admit. You'll generally find Robert in the ETF aisle, when he's not serving as director of the Computer Security Institute (CSI) . Wait, was this bio bit supposed to be humorous? Crap.