Accuvant Labs' Charlie Miller talks to Tom's Hardware about the Defender's Dilemma, the security of data in the cloud, looking for vulnerabilities in notebook batteries, and the ramifications of using Apple's iPad in an enterprise environment.

If you're not already familiar with Charlie Miller, check out Behind Pwn2Own: Exclusive Interview With Charlie Miller and Hacking The iPhone, iPod, And iPad With A Web Page, two of our previous interviews with him.

Alan: Hi Charlie, thanks again for taking the time to sit down and talk with me and the readers of Tom’s Hardware. I know how busy it gets around Black Hat.

Charlie: Yep, between my "day job" doing consulting and writing slides and finishing research, this is my second busiest time of the year. The only worse time is right before Pwn2Own! I can't wait until Black Hat and DEF CON are finished to relax and start some new research.

Alan: A lot has changed since the last time we chatted. The impact and critical importance of computing security has really just begun to be appreciated by mainstream users. The New York Times had a great feature on the development of Stuxnet and detailed how the good guys undermined and crippled Iran’s nuclear program. We’ve seen the bad guys attack Lockheed Martin through a targeted effort that began with compromising RSA SecurID. Lastly, we’ve seen end-users directly impacted by the actions of groups like Anonymous and LulzSec. It’s the good, the bad, and the ugly.

So for today, I’m hoping I can get your thoughts on some big-picture stuff before I pick your brain on the new iOS jailbreak and the battery firmware vulnerability.

Charlie: Well, who the bad guys and who the good guys are can be a bit hard to determine sometimes. I'm sure the Iranians don't consider Stuxnet to be a force for good. But yes, let's talk about the big picture.  

Alan: Well, I know I'm a good guy, and you're a good guy. People can trust us. We're doctors. Anyway, after the Brighton Bombing in ’84, the IRA released a statement that included the line "...remember we only have to be lucky once. You have to be lucky always." When it comes to computing security, it seems like it’s the same challenge. Only in this case, it’s even worse. The bad guys are coming from multiple fronts. You have targeted attacks, automated botnets, and broad social engineering spam. You also have different motives ranging from espionage and financial or political gain to activists looking to make a statement. While there was a political process that could bring peace to the United Kingdom, you’re not going to be able to negotiate with someone looking to steal credit card info or sensitive data.

Can we actually win this war or are we just hoping to minimize our losses?

Charlie: Yes, we call it the Defender's Dilemma. Defense is always harder because you have to be perfect, where attackers only have to find one flaw. This is why it’s so much more fun to attack Apple than to work for Apple!

I have to say, things are a bit bleak when you put it that way. There will always be vulnerabilities and there will always be criminals, so it’s hard to figure the way out. Especially as end users there is almost nothing you can do; you have to rely on the security of the software you run and have little control over how secure it is. As a society, we cannot eliminate computer attacks. However, what we can do (and this is the approach the industry is sort of taking) is make it so hard and expensive to pull off attacks that it becomes economically infeasible for most attackers. And even for those with the expertise to still pull off the attack, it minimizes the number of attacks they can perform. The way we make it more difficult is to reduce the number of vulnerabilities and ensure users’ software is up to date and "secure by default”. Also, make the OS resilient to attack with things like stack canaries, ASLR, DEP, and sandbox applications so that multiple exploits are needed. We also need to better control the software loaded on our devices (i.e. Apple's App Store model). So, instead of having to write a single exploit, it takes three or four in order to perform an attack. This means most attackers won't be able to pull it off, and those who can will have to spend much more time working it out.