Is government on the wrong road with cybersecurity?
- By Amber Corrin
- May 21, 2013
NSA's Chris Inglis offered a recipe for heartbreak at a CSIS event. (NSA photo)
The calls to improve cybersecurity grow louder by the day, but how can it work if the approach is wrong?
The proliferation of networks, devices and access, combined with rapidly growing architecture and social networking, is fueling a landscape rife with cyber crime and burgeoning threats. The government is working to roll out a number of measures to boost cybersecurity, but experts say efforts must weave in new models and build on traditional tactics.
"It's almost impossible to achieve a static advantage in cyberspace – whether that's a competitive advantage or a security advantage – when things change every minute of every hour of every day. And it's not just the technology that changes; it's the employment of that technology; the operations and practices," Chris Inglis, National Security Agency deputy director, said May 21 at the Center for Strategic and International Studies in Washington. "If your security depends upon a static advantage and the static nature of compliance-based standards, your heart's going to be broken on a fairly regular basis."
Getting the government on the right cybersecurity track undoubtedly means making major changes, but it is difficult to get the ball rolling amid deep, systemic flaws, speakers at CSIS indicated. Much is riding on the latest efforts – in particular, information-sharing and collaboration – but those are layered atop outdated models that also must change with the times.
"We've been conditioned to build defense in depth as an architectural model, and that architectural model today is so easily evaded" despite talent and focus going into cybersecurity, said David DeWalt, chairman and CEO of FireEye, a computer security firm. "We have a significant deficit in our defense models. Probably the most interesting place we're seeing right now is the dislocation between offense and defense. I've never seen the gap be wider than it is today."
That broken defense model contributes to what has become a $1 trillion epidemic of cyber attacks, DeWalt said. Among the statistics he rattled off: An average U.S. business is hit with an attack 100 times a day; 9,000 malicious websites are created worldwide every day; and 95 percent of U.S. companies have their computers compromised every day.
"At the end of the day we need to do more than simply take the slings and arrows that come our way, going into a fetal crouch," Inglis admitted. The U.S. community of cyber defenders "by and large is increasingly aware of its vulnerabilities and the threats against it, but is not yet...integrated or collaborative enough across the multiple parties in order to prevail."
Government-directed measures to improve information-sharing and collaboration are critical, but they are still premature – including limitations in this year's cybersecurity executive order and in pending legislation that still need to be resolved, Inglis noted.
"If convergence is a reality in cyberspace, then integration in our engineering and collaboration – the human form of integration – must be a necessary response," he said. "We have to put all of our efforts together in order to make this a more defensible, more resilient set of networks, and to then actually defend them."
Information-sharing is a major part of that, although that term is over-used and broad, said Jenny Menna, Homeland Security Department director of stakeholder engagement and critical infrastructure resilience. It also means more than just sharing specific threat intelligence between agencies and companies.
"Sometimes that's actionable indicators...sometimes that's sitting down with CEOs or CIOs to make sure they really understand the threat landscape, what are the most important things where they should allocate scarce resources, what are the decisions they're making that may introduce significant risk, and what are the new technologies out there" that are worthwhile, Menna said.
Beyond sharing information and collaboration, implementation remains a crippling problem, others noted. Applying existing security measures – such as critical security controls that have been identified by entities like the SANS Institute, NSA and the Australian government, for example –would seem obvious, but agencies and companies still fail to do so, leaving them vulnerable.
"We know what we need to do, we just don't do a very good job of implementing," said John Gilligan, president of the Gilligan Group and former Air Force CIO. "As CIO I learned it doesn't cost a lot of money to implement these baseline controls...because most of them are essential to operating and managing the network. It's just doing them in a disciplined manner. So step one is implement the baseline critical security controls."
In other words, one of cybersecurity's biggest hurdles is people – not technology.
"It is humans, after all, who manage and conduct the affairs known as security practices protocols," Inglis pointed out. "The reason adversaries are so enormously successful against our networks today is that it's our implementation that's at issue. Therefore we need to focus on the behavior of humans, not just the behavior of technology."