What to Expect from the NSA Hacker Turned White House Cyber Advisor
The choice of Rob Joyce, former head of the National Security Agency’s Tailored Access Operations unit as cyber security coordinator puts an experienced offensive cyber operator at the nexus of the nation’s cyber policy and strategy at a time when nation-state cyber interference is at the forefront of public consciousness.
Joyce succeeds Michael Daniel, who had a public policy, economist and finance background and spent nearly a decade in cyber policy at the Office of Management and Budget and the White House. Joyce’s background, by contrast, is as an operator in the cyber realm, bringing an intimate understanding of the threat to the forefront of national cyber policy.
As cyber coordinator, Joyce is not the federal chief information security officer (CISO). That post is largely focused on securing the federal enterprise; the cyber coordinator drives policy beyond the federal government. “The cyber coordinator is also interested in cybersecurity across the entire digital ecosystem,” including private industry, state and local governments and foreign governments, as well. “So it’s a much broader role than what the federal CISO focuses on,” says Daniel, who is now president of the Cyber Threat Alliance, a non-profit focused on cyber threat sharing across the industry. “There is some degree of overlap and complementarity – obviously the cybersecurity coordinator has to care about the security of federal networks – but the cybersecurity coordinator has a broader mandate than that.”
Little is publicly known about NSA’s offensive cyber activities. But in a rare public appearance last August at the USENIX 2016 conference, Joyce described the five steps to a successful cyber intrusion – initial exploitation, establish presence, install tools, move laterally and collect/ex-filtrate/exploit – and then walked through the weaknesses he and his hackers came across and exploited each day.
“If you really want to protect your network,” he said then, “you really have to know your network. You have to know the devices, the security technologies, and the things inside it.” His clear message: His team often knew better than the network’s managers. Indeed, while NSA hackers might not understand products and technologies as well as the people who design them, Joyce said they learn to understand the security aspects of those products and technologies better than the people who created them.
“You know the technologies you intended to use in that network,” he said. “We know the technologies that are actually in use in that network. [There’s a] subtle difference. … You’d be surprised at the things that are running on a network versus the things you think are supposed to be there.”
Penetration-testing is essential, as is follow-up. Joyce’s OTA regularly conducted Red Team testing against government networks. “We’ll inevitably find things that are misconfigured, things that shouldn’t be set up within that network, holes and flaws,” he said. The unit reported its findings, telling the network owner what to fix.
Then a few years later, it would be time to test that network again. “It is not uncommon for us to find the same security flaws that were in the original report,” Joyce said. “Inexcusable, inconceivable, but returning a couple of years later, the same vulnerabilities continue to exist. I’ve seen it in the corporate sector too. I’ve seen it in our targets.”
Laziness is a risk factor all its own. “People tell you you’re vulnerable in a space, close it down and lock it down,” Joyce said, reflecting on the fact that network administrators frequently don’t take all threats and risks seriously enough. “Don’t assume a crack is too small to be noted or too small to be exploited. … There’s a reason it’s called advanced persistent threats: Because we’ll poke and we’ll poke and we’ll wait and we’ll wait and we’ll wait, because we’re looking for that opportunity to [get in and] finish the mission.”
As an offensive cyber practitioner, Joyce sought to identify and, when needed, exploit the seams in government and enemy networks. He focused on the sometimes amorphous boundaries where the crack in the security picture might come from getting inside a personal device, an unsecured piece of operational security, such as a security camera or a network-enabled air conditioning system, or even an application in the cloud. “Cloud computing is really just another name for somebody else’s computer,” he said. “If you have your data in the cloud, you are trusting your security protocols – the physical security and all of the other elements of trust – to an outside entity.”
Most networks are well protected, at least on the surface. They have high “castle walls and a hard crusty shell,” he said. But “inside there’s a soft gooey core.”
Figuring out how to protect that core from a national security and policy perspective will be Joyce’s new focus, and if Daniel’s experience is any indicator, it will be a challenge.
From his perspective, cybersecurity is only partly about technology. “Adversaries tend to get into networks through known, fixable vulnerabilities,” Daniel says. “So the reason those vulnerabilities still exist is not a technical problem – because we know how to fix it – it’s an incentive problem – an economics problem.” That is, network owners either fail to recognize the full extent of the risks they face or, if they do, may be willing to accept those risks rather than invest in mitigating them.
The challenge, then, is formulating policy in an environment in which the true level of risk is not generally understood. In that sense, Joyce’s ability to communicate the extent to which hackers can exploit weaknesses could be valuable in elevating cyber awareness throughout the White House.
“The NSC is about managing the policy process for the national security issues affecting the US government,” Daniel explains. “You don’t have any direct formal authority over anyone. But you do have the power to convene. You have the power to raise issues to people in the White House. You have the ability to try to persuade and cajole. The background he brings will obviously color what he prioritizes and what he puts his time against. But the role itself will not be dramatically different. … understanding how to get decisions keyed up in a way that you can actually get them approved.”
Joyce’s background could affect how this administration views commercial technologies, such as cloud services, mobile technology and other advances that, while ubiquitous in our daily lives, are not yet standard across the federal government.
“Trust boundaries now extended to partners,” Joyce said a year ago. “Personal devices – you’re trusting those on to the network…. So what are you doing to really shore up the trust boundary around the things you absolutely must defend? That for me is what it comes down to: Do you really know what the keys to the kingdom are that you must defend?”
National security cyber policy is not just defensive, however, and having a coordinator with a keen insider’s understanding of offensive cyber capabilities could have a significant long-term impact on national cyber strategy.
Just as Daniel sees cybersecurity as an incentives, or economics problem, Kevin Mandia, chief executive at the cyber security firm FireEye and founder of Mandiant, its breach-prevention and mitigation arm, sees incentives and disincentives as playing a critical role for cyber criminals and nation-state attackers, alike. Simply put, he says, the risk-reward ratio tilts in their favor, because the consequences of an attack do not inflict enough pain.
Mandia agrees that the first priority for U.S. cyber policy should be self-defense. “Every U.S. citizen believes the government has a responsibility to defend itself,” he said at the FireEye Government Forum March 15. “So first and foremost, our mission security folks must defend our networks. But the second thing the private sector wants is deterrence. We need deterrence for cyber activities.”
And in order to develop an effective deterrence policy, he argues, the nation needs fast, reliable attribution – the ability to unequivocally identify who is responsible for a cyber attack.
“I’d take nothing off the table to make sure we have positive attribution on every single cyber attack that happens against U.S. resources,” Mandia says. “Because you can’t deter unless you know who did it. You have to have proportional response alternatives, and you have to know where to direct that proportionate response.”
Where Joyce stands on deterrence and attribution is not yet clear, but what is clear is that sealing off the cracks in federal network security is sure to get more intense.
“A lot of people think the nation states are running on this engine of zero-days,” Joyce said a year ago, referring to unreported, unpatched vulnerabilities. “It’s not that. Take any large network and I will tell you that persistence and focus will get you in, will achieve that exploitation without the zero days. There are so many more vectors that are easier, less risky and quite often more productive than going down that route.”
Closing off those vectors forces threat actors to assume more risk, expose zero-day exploits and operate with less cover. When that happens, the balance of cyber power could finally start to tilt away from the hackers.