As the year is coming to a close, news headlines were dominated by reports of high-profile security attacks, some launched by “hacktivists” such as Anonymous and LULZSEC.
But something larger was brewing. Amidst hacktivists’ attacks on Sony, HBGary and NATO, highly sophisticated, clandestine attackers—the kind with the rarefied expertise, deep pockets and specialized resources typically only seen in nation-state adversaries—were actively infiltrating a broad range of targets.
These attacks were different: they were patient, stealthy and leveraged a potent combination of technical skill and social savvy. Some used clever social engineering to get a foothold into their target organizations, while others used zero-day vulnerabilities—previously unknown holes in software—to penetrate defenses.
While advanced attacks have happened for years, IT security experts observed recent attacks had grown bolder and more frequent. Recent attacks were also highly targeted, customized, well-researched and, in many cases, employed both technical and social
The term used to describe such complex, sophisticated attacks was
“advanced persistent threats” (APTs), but as IT security experts quickly pointed out, APTs were only as advanced as they needed to be to get the job done. A concrete definition is elusive and, as cautioned, “Defining it could limit us and lead us to be blindsided. We need to constantly revisit the characteristics because they’re always changing.”
Much of the day’s focus was on the techniques of highly organized attackers. such
advanced threats, which include APTs, span from corporate espionage to hacktivism.
This article distills certain key insights from those discussions and
aspires to advance the industry’s dialog on advanced threats, spur disruptive innovation and disseminate some of our learnings from some of the most seasoned professionals in information security.
From a Cookie-Cutter Approach to Adaptive
In 2000, the I LOVE YOU worm crippled more than 50 million Pcs. The delivery mechanism was simple but effective: an e-mail showed up in your in-box with a subject line of “iloveyou.” When people clicked on the e-mail’s attachment, titled “love-leTTer-Foryou,” they were infected with a computer worm. while the damage was significant, a
partial solution to this problem came in the form of antivirus software: a signature could be deployed to antivirus agents that would identify the file as malicious and arrest its actions.
Today, generic malware is still profuse but signature-based defenses, at either the network or host layer, can greatly decrease the odds of infection. What makes recent
advanced threats different is their defiance of a signature. In the world of advanced threats, malware evolves quickly, and security experts have described several cases of special-purpose malware custom-developed specifically for their targets. Some were
compiled within hours of launching the attack.
It became clear that enterprises targeted by highly organized attackers cannot depend on signature-based “bread and butter” security tools as a sole means of defence. While the payloads of some advanced threats were fairly standard, entry strategies were often custom tailored.
Attackers typically used social networking sites to gather intelligence and identify specific users within an organization. Some of the main infection vectors that cited were e-mail, Skype and instant messages with malware payloads in the form of PDFs, compressed HTML, script files, executables and attachments.
customization of attack techniques extend through data exfiltration.
Advanced threats often use sophisticated methods for compressing, encrypting and transmitting data to other compromised organizations, leaving little evidence of the origin of the attack or the destination for stolen information. This move from generic to tailored, from cookie-cutter to adaptive, means that security organizations need to think beyond signatures and re-evaluate how effective their current defenses are.
Remember that people, not technology, were the Achilles heel in most defensive strategies.
People are the Weakest Link
“People are the weakest link” is perhaps the biggest cliché in information security. Security experts have long understood that users make bad choices, click on links they shouldn’t and install malware through simple ruses.
Corporate IT departments deploy multiple controls to help deal with this threat: e-mail filtering solutions catch many attacks before they make it to users, malicious links are blocked by the network, network scanners look for malicious content, and host-based antivirus (the last line of defense) tries to stop what slips through the cracks.
This process works well for generic, shotgun attacks in that signatures can be updated quickly to immunize users. Advanced attackers, however, are now creating highly credible scenarios in which they convince users to click on dialog boxes warning of fake software updates, retrieve content from quarantined areas and act (unknowingly) on behalf of the attacker.
Attackers have become dangerously adroit at using our weaknesses and behaviors against us. Attackers are creatively leveraging people inside the company to help accomplish their goals. “Internet scams are supposed to be sloppy, but they work.”
Advanced threats defy that stereotype. Experts put a fine point on it: “The perimeter is not a firewall; it’s our users. They don’t treat their computer as an adversary; they treat it as a tool—an extension of themselves—so they don’t question what it tells them.”
Addressing the people problem will take more than technology. Organizations need to
drive a sense of personal responsibility for security among employees.
Attackers Aim for Advantage, Not Infamy
Advanced attacks are typically not the product of hobbyists. These attacks often require months of planning, mapping out internal networks by looking at the fringes.
The reconnaissance can go much further: targeting key employees, deconstructing their life by scouring social media, custom-crafting an attack so that it is stealthy, patient, and very effective.
Cybercriminals, the ones who look to steal credit card numbers and other
commoditized and sellable data, have become increasingly sophisticated but advanced
attacks are different. Increasingly, they focus on espionage—stealing specialized data
that may be of high value and strategic importance to the commissioning entity, which
can be foreign governments, rival corporations and organized crime groups. The entities behind advanced attacks literally mean business.
Also, entities perpetrating many advanced attacks are substantively different from the
hacktivists groups that have attracted attention in recent times. Hacktivists want to
embarrass and expose their targets’ activities, taking pride in publishing their conquests.
Many advanced attackers, in contrast, have the goal of stealth. They do not want to be
discovered or seek publicity.
Now some advanced threats are now masquerading as hacktivist attacks, with the goal being to confuse forensics and place blame on groups that are often eager to accept it. This pattern makes it difficult to size the scale of advanced threats: a willing scapegoat makes post-incident attribution particularly problematic.
The New Normal: Act as Though You Are Already Hacked
The events of the year have shown that determined adversaries can always find exploits through people and in complex IT environments. It’s not realistic to keep
adversaries out. Organizations should plan and act as though they have already been breached.
Three foundational principles of security are compartmentalization, defense in depth and least privilege. in combination, these three tenets dictate that if one system (or person) is compromised, it should not result in a compromise of the entire system.
While simple in concept, these tenets have proven complicated to implement in practice. Organizations have long relied on the notion of a “perimeter,” where a big thick wall—in the form of firewalls and gateway defenses—guards the organization, with good guys (insiders) on one side of the wall and attackers on the other.
Security perimeters are now considered a construct of the past. Boundaries are nearly
impossible to define in modern organizations. The inclusion of partially trusted users
such as customers, suppliers, contractors, service providers, cloud vendors and others
have made organization boundaries very porous. Beyond the eradication of traditional
organizational boundaries, the consumerization of IT has brought a rash of unmanaged
devices into the enterprise and exposed the organization to services (and suppliers) that are opaque.
IT consumerization has also blurred the line between the business lives and
the personal lives of employees. We have moved from the illusion of a perimeter-driven defense to living in a state of compromise.
Accepting that some systems, some people, and some services may already be under the control of attackers changes information security strategy. it forces a return to the core principles of compartmentalization, defense-in-depth, and least privilege.
Organizations need to focus on closing the exposure window and limiting damage through efforts to compartmentalize systems, stop sensitive data egress and contain malfeasance. This new model also demands that we rethink old habits of sharing sensitive corporate information—such as source code, product plans and strategic roadmaps—using collaborative processes that presume perimeter defenses can keep attackers out.
Security improves through greater situational awareness: gaining the ability to
understand what’s happening beyond our network boundaries to detect threats on the horizon. Organizations get smarter by looking beyond their infrastructure and observing the ecosystem. The ecosystem approach to security relies on organizations actively sharing information with other organizations about threats. It also demands greater visibility into the security of suppliers and service providers within one’s supply chain.
The key is to know what digital assets are important to protect, where they reside, who
has access to them and how to lock them down in the event of a breach. This ability to
tighten the net before and during an attack is key, and it requires a mature process for
incident handling. Incident response should not be considered exclusively a security
function. Instead, it is an organizational competency that must be developed and
continually honed well before an attack occurs. if organizations are planning responses
as an attack unfolds, they are too late. A competency approach allows remediation
activities to kick in automatically—like a reflex.
The Road Ahead
The reality of advanced threats demands a disruptive approach to defense—one where
enterprises can be agile and thrive in a contested environment. This approach must be
applied holistically: approaching advanced threat defense not as a discrete function but
as a natural consequence of robust but agile security.
Many of the holes that exist today come from an unmanageably complex iT infrastructure. Given that information security is a “weakest link” problem, only through understanding our assets, processes and endpoints do we have a chance at effective defense. Unraveling complexity and fielding a successful defense means that we also need to think creatively about the range of attacker motivations, which can extend far beyond data theft.
With every new technology, we have the ability to weave security into its fabric, to begin anew. We are at the start of an industry-wide move to cloud-based services and systems. We stand on the precipice of a sea-change in technology. There is a new mantra that goes within the industry saying “If we can’t get it right with cloud, shame on us.”
Today more than ever, security is an ecosystem problem in which every constituent has a responsibility. Attackers are collaborating, sharing information, going after the supply chain, co-opting careless insiders and evading our long relied-upon defenses. we need disruptive collaboration and innovation in defense. Through collaboration, information sharing and increasing our agility, we can successfully fend off APTs and other advanced threats.
Happy Holidays! And a blessed new year to all!