Cloud migration and remote working attracting new breed of cyber-threat

The malicious programmes, known collectively as Highly Evasive Adaptive Threats (HEAT) attacks are targeting web browsers and employ techniques to evade detection by multiple layers in current security stacks including firewalls, Secure Web Gateways, sandbox analysis, URL Reputation, and phishing detection.

HEAT attacks are being used to deliver malware or to compromise credentials, that in many cases leads to ransomware attacks.

Now, flags have been raised thanks to analysis of almost 500,000 malicious domains, carried out by Menlo Security Labs. The research team at Menlo discovered that 69% of the websites studied used HEAT tactics to deliver malware.

These attacks allow bad actors to deliver malicious content to the endpoint by adapting to the targeted environment. Since July 2021, Menlo Security has seen a 224% increase in attacks of this nature.

Amir Ben-Efraim, co-founder and CEO of Menlo Security, said:

“With the abrupt move to remote working in 2020, every organisation had to pivot to a work from an anywhere model and accelerate their migration to cloud-based applications. An industry report found that 75% of the working day is spent in a web browser, which has quickly become the primary attack surface for threat actors, ransomware and other attacks.

“The industry has seen an explosion in the number and sophistication of these highly evasive attacks and most businesses are unprepared and lack the resources to prevent them,” Ben-Efraim added.

“Cyber Threats are a mainstream problem and a boardroom issue that should be on everyone’s agenda. The threat landscape is constantly evolving, ransomware is more persistent than ever before, and HEAT attacks have rendered traditional security solutions ineffective,” Ben-Efraim continued.

HEAT attacks leverage one or more of the following core techniques that bypass legacy network security defences:

• Evades Both Static and Dynamic Content Inspection: HEAT attacks evade both signature and behavioural analysis engines to deliver malicious payloads to the victim using innovative techniques such as HTML Smuggling. This technique is used by threat actors including Nobelium, the hacking group behind the SolarWinds ransomware attack. In one recent case, dubbed ISOMorph, the Menlo Labs research team observed the campaign using the popular Discord messaging app to host malicious payloads.
• Evades Malicious Link Analysis: These threats evade malicious link analysis engines traditionally implemented in the email path where links can be analysed before arriving at the user.
• Evades Offline Categorisation and Threat Detection: HEAT attacks evade web categorisation by delivering malware from benign websites, either by compromising them, or patiently creating new ones, referred to as Good2Bad websites. One active threat campaign dubbed SolarMarker, employs SEO poisoning. The campaign started by compromising a large set of low-popularity websites that had been categorised as benign, infecting these websites with malicious content.
• Evades HTTP Traffic Inspection: In a HEAT attack, malicious content such as browser exploits, crypto-mining code, phishing kit code and images impersonating known brand logos is generated by JavaScript in the browser by its rendering engine, making any detection technique useless.

The top three brands impersonated in phishing attacks are Microsoft, PayPal, and Amazon. A new phishing website imitating one of these brands is created every 1.7 minutes.

John Grady, ESG Senior Analyst, said:

“Highly Evasive Adaptive Threat (HEAT) attacks evade existing security defences by understanding all the technology integrated into the existing security stack and building delivery mechanisms to evade detection.

“Organisations should focus on three key tenets to limit their susceptibility to these types of attacks: shifting from a detection to a prevention mindset, stopping threats before they hit the endpoint, and incorporating advanced anti-phishing and isolation capabilities,” Grady advised.