Accepting The Inevitability Of Attack

Protecting against attacks should not be the sole focus of an enterprise's security strategy, especially because attacks are a given today, according to experts who will debate the future of security tomorrow at the Forrester Security Forum in Boston.

"Security has been sold a bill of goods by vendors that protection is what we need. But protection is only one part of it," says Andrew Jaquith, senior analyst for Forrester Research, who will chair the "Looking Forward to Getting Attacked" panel at the Forrester event.

Jaquith says security has three elements: prevention, detection, and response. "If you put all of your eggs in the prevention basket, you find a surprise when those protections fail," says Jaquith, who, along with panelists Herbert H. "Hugh" Thompson, chief security strategist for People Security, and Daniel Geer, chief scientist emeritus at Verdasys, spoke with Dark Reading in advance of the conference.

Instead of just trying to stop everything at the "front door," he says, survivability and recoverability should be what's emphasized. "These things matter a lot," Jaquith says. "Customers tell us endpoint security doesn't work like it used to. But when you set the bar at absolute perfection, of course you're going to be disappointed."

People Security's Thompson says it's not a question of how the security industry has gone wrong, but instead that this is where it's going. "When it comes to security, we're used to thinking in absolutes: We're secure, we're not secure," Thompson says. "But those of us who've worked in security for a long time know it's a continuum. We know attackers are so sophisticated and are evolving techniques faster than we can because their business model is more efficient than ours.

"So how do we focus on recovery and survivability? How do we first notice we've been attacked, and how do we take reasonable precautions?" he says. It's more about mitigating an attack that's likely to happen, he says.

Verdasys' Geer points to this year's Verizon Business breach report, which showed that the big breaches didn't target patchable vulnerabilities in software, a shift from its report last year. "So if big breaches are unpatchable, why do we care about patching?" Geer says.

The underlying problem in security is its lack of agility in shifting with the threats, according to Geer. "Our problem is a lack of agility and the absence of demand for it," he says.

With the mobile revolution and social networking, more security decisions are being placed in the end user's hands; at the same time, social engineering is becoming more of a problem for organizations.

Forrester's Jaquith predicts that attacks will target people more than devices: "A lot of traditional vectors don't work as well other than the odd exploit," he says. "So attackers will have to rely more on social engineering techniques than a buffer overflow to take over an entire system ... The challenge is getting users to do the right thing."

That's a tall order, Thompson maintains. "Users have shown consistently that they make bad security choices with technology," he says. "You've got to wonder if the real threat in two or five years is the personalized targeted attack -- not the viruses out there with a defined signature that my AV can catch."

Attackers could glean enough information from a user's social networking posts to fool that person into opening a file or clicking on a link, or to simulate that person, Thompson says. "At a point, I would fall for something if someone had set up the right context -- if the email came at the right time with the right key words and references," he says.

Thompson says he'd like to see tools that help users make better decisions in these situations -- to provide warning signals, for instance: "You posted this information on Facebook, LinkedIn" so that information about a dinner, for example, was available on those sites. That way the email citing meeting with someone you don't remember could be suspicious.

Mobile vendors will do a better job building security into their operating systems and shielding users from basic attacks, notes Forrester's Jaquith. "There have been shrill blog posts [warning] about mobile security threats. But most of those threats are about what you [the user] decide ... and are social" engineering-type attacks, he says.

Jaquith says a new breed of mobile security vendors will provide different kinds of products and services than we have today in security. "Today's model is on the endpoint, with an agent piece of software that provides services on your behalf. But new platforms are much closer to sealed boxes, more like toasters than PCs ... no one wants a PC AV model replicated on mobile devices."

Security tools for mobile devices will center around theft and data loss prevention, data resiliency, and backup, he says.

But Thompson says he wonders if anyone really knows what threats mobile devices eventually will face. "The smartphone is more and more of a gateway into the enterprise than it was in the past. We're also seeing a convergence of platforms, and as these devices manage more and more sensitive information, it seems the economics of an attack would be easier ... through a mobile device," he says. "There's going to be a new type of threat that evolves out there, and I don't think mobile security vendors know what that threat is going to look like [yet]," he says.

Have a comment on this story? Please click "Discuss" below. If you'd like to contact Dark Reading's editors directly, send us a message.