The weak links in an increasingly dynamic threat landscape

The Cisco 2014 Midyear Security Report, released today at Black Hat, examines the “weak links” in organizations – outdated software, bad code, abandoned digital properties, or user errors – that contribute to the adversary’s ability to exploit vulnerabilities with methods such as DNS queries, exploit kits, amplification attacks, point-of-sale (POS) system compromise, malvertising, ransomware, infiltration of encryption protocols, social engineering and “life event” spam.

The report also shows that focus on only high-profile vulnerabilities rather than on high-impact, common and stealthy threats puts these organizations at greater risk. By proliferating attacks against low-profile legacy applications and infrastructure with known weaknesses, malicious actors are able to escape detection as security teams focus instead on boldface vulnerabilities, such as Heartbleed.

Researchers closely examined 16 large multinational organizations, who, as of 2013, collectively controlled over $4 trillion in assets with revenues in excess of $300 billion.

This analysis yielded three security insights tying enterprises to malicious traffic:

• “Man-in-the-Browser” attacks pose a risk for enterprises: Nearly 94 percent of customer networks observed in 2014 have been identified as having traffic going to websites that host malware. Specifically, issuing DNS requests for hostnames where the IP address to which the hostname resolves is reported to be associated with the distribution of Palevo, SpyEye, and Zeus malware families that incorporate man-in-the-browser (MiTB) functionality.
• Botnet hide and seek: Nearly 70 percent of networks were identified as issuing DNS queries for Dynamic DNS Domains. This shows evidence of networks misused or compromised with botnets using DDNS to alter their IP address to avoid detection/blacklist. Few legitimate outbound connection attempts from enterprises would seek dynamic DNS domains apart from outbound C&C callbacks looking to disguise the location of their botnet.
• Encrypting stolen data: Nearly 44 percent of customer networks observed in 2014 have been identified as issuing DNS requests for sites and domains with devices that provide encrypted channel services, used by malicious actors to cover their tracks by exfiltrating data using encrypted channels to avoid detection like VPN, SSH, SFTP, FTP, and FTPS.

The researchers also found that the number of exploit kits has dropped by 87 percent since the alleged creator of the widely popular Blackhole exploit kit was arrested last year. Several exploit kits observed in the first half of 2014 were trying to move in on territory once dominated by the Blackhole exploit kit, but a clear leader has yet to emerge.

Java continues its dubious distinction as the programming language most exploited by malicious actors. Cisco security researchers found that Java exploits rose to 93 percent of all indicators of compromise (IOCs) as of May 2014.

There were also unusual upticks in malware within vertical markets. For the first half of 2014, the pharmaceutical and chemical industry, a high-profit vertical, once again placed in the top three high-risk verticals for Web malware encounters. Media and publishing led the industry verticals posting nearly four times the median Web malware encounters, and aviation slid into third place with over twice the median Web malware encounters globally.

The top most affected verticals by region were media and publishing in the Americas; food and beverage in EMEA (Africa, Europe and the Middle East) and insurance in APJC (Asia-Pacific, China, Japan and India).