공격자의 선제 공격 우위 정량화

Tenable Research has just released a report on the difference in time between when an exploit is publicly available for a given vulnerability and the first time that a vulnerability is assessed.

For this study, we analyzed the 50 most prevalent critical and high-severity vulnerabilities from just under 200,000 vulnerability assessment scans over a three-month period in late 2017 to anchor the analysis to the real world. We used these vulnerabilities to derive the “time to exploit availability” and “time to assess” to calculate the median delta.

The delta represents the first move that both the defender and attacker can make in this race. While it does not represent the full OODA Loop (observe, orient, decide, and act) for both parties, it does indicate who has the best start – and who will ultimately win.

A negative delta means attackers have a window of opportunity to exploit a vulnerability before defenders are even aware that they are at risk.

Attackers are racing ahead

Our analysis shows that the median delta was -7.3 days. The median time to exploit was 5.5 days, compared to a median time to assess of 12.8 days. On average, this gives attackers a seven-day head start on the defenders.

The delta was negative for 76 percent of analyzed vulnerabilities. So, on a vulnerability-by-vulnerability basis, the attackers seize the first-mover advantage more often than not.

When the delta was positive, it was usually because it took so long for an exploit to become available – rather than the defenders’ speedy scanning frequency. The fact that for 34 percent of the analyzed vulnerabilities, an exploit was available on the same day the vulnerability was disclosed is sobering. But it really gets interesting when we drill down into the individual vulnerabilities.

Twenty-four percent of the 50 most prevalent vulnerabilities we analyzed are actively being exploited in the wild by malware, ransomware or exploit kits. A further 14 percent were sufficiently critical to be discussed in the media. The sample set contained vulnerabilities being targeted by the Disdain and Terror exploit kits, Cerber and StorageCrypt ransomware and even by APT groups such as Black Oasis to install the FinSpy surveillance software.

Seizing the advantage back

Most security professionals have a hunch they’re playing catch up, but having the attackers’ lead quantified helps to determine how far behind we are and what we need to do to close the gap.

Many organizations conduct vulnerability assessments on a monthly or even quarterly cycle. This is mostly driven by internal factors, rather than external drivers, such as monthly patch cycles set by the company. We tend to forget that we have an adversary who is dictating the rules of engagement. While we have no control over when the attacker decides to attack, or how, we do have control over our own environment.

Our own analysis of scanning behavior indicates that just over 25 percent of organizations are conducting vulnerability assessments with a frequency of two days or fewer. This is an achievable goal, and will reduce the head start that attackers have for most vulnerabilities. But, with latency inherently built into the process, this still leaves a residual risk that cannot be solved solely by improving the scan frequency.

A more effective approach to vulnerability and Cyber Exposure management is not based on a start-stop model or discrete cycles. Instead, it must be based on:

• Continuous assessment of our security posture
• Proactively addressing predictable risks
• Rapidly reacting to unanticipated and emerging ones

The threat landscape is evolving at an unprecedented pace across an expanding attack surface, necessitating a more agile process that functions as a feedback loop. The new emerging discipline of SecDevOps already provides some established best practices. But this will also require that security and operational teams are better aligned and cooperate more closely.

We shouldn’t underestimate the value of vulnerability assessment in providing intelligence and situational awareness of Cyber Exposure and providing the “feedback” in the loop. Even if it isn’t feasible to remediate every vulnerability at once, continuous assessment can drive ad hoc, emergency response remediation or determine that mitigating controls such as access control must be applied to bridge the remediation delay and resulting exposure gap.

Speed up the time to assess vulnerabilities

For recommendations on how you can speed up time to assess vulnerabilities and detailed insights into our analysis, download the Quantifying the Attacker’s First-Mover Advantage report.