Can Hackers Bypass Multi-Factor Authentication Using Cookies?

Article Highlights
Off On

Researchers at Varonis Threat Labs have uncovered an intricate method allowing cyber attackers to get around multi-factor authentication (MFA) and gain lasting access to enterprise cloud environments. By exploiting browser cookies, attackers can completely evade traditional security measures and remain undetected. This research, revealed in April, highlights the growing sophistication of session hijacking techniques, combining browser extensions and scripts to steal authentication tokens. It underscores a troubling evolution, indicating that even advanced security setups like MFA may be insufficient against such intricate attacks.

The Core Mechanics of Cookie-Based Attacks

At the heart of this attack are stolen session cookies, small pieces of data stored in a user’s browser to confirm previous authentication, including MFA. Once these cookies are captured, attackers can impersonate actual users, gaining undetected access to services like Microsoft 365, Google Workspace, and Amazon Web Services (AWS). This method bypasses security alerts, making it exceptionally dangerous.

Varonis researchers have demonstrated a proof-of-concept: a custom Chrome browser extension paired with PowerShell scripts to extract Azure Entra ID session cookies. Identified as ESTSAUTH and ESTSAUTHPERSISTENT, these cookies are obtained anytime a user logs into Microsoft’s authentication portal. After being extracted and inserted into an attacker’s browser, these cookies grant immediate access to the victim’s cloud services, effectively bypassing MFA prompts. The malicious extension operates by monitoring login events to domains such as login.microsoftonline.com, capturing authentication cookies in real time, and uploading them to an attacker-controlled endpoint like a Google Form. This stealthy approach circumvents typical malware behaviors, making endpoint detection exceedingly challenging.

Advanced Techniques and Marketplace for Stolen Data

On Windows systems, cookies are encrypted using the Data Protection API (DPAPI), which ties encryption keys to user profiles and specific machines. However, infostealers can exploit decrypted cookies stored in the process memory or extract encryption keys depending on their access level. Capturing these cookies may involve adversary-in-the-middle (AITM) attacks using reverse proxy tools and malicious extensions requesting high-level permissions to access session data directly. Stolen session data is circulated within a Malware-as-a-Service (MaaS) ecosystem. Infostealers distribute malware widely to collect credentials, tokens, and cookies, which are then sold on darknet markets to buyers such as ransomware operators and initial access brokers. Particularly, cookies that provide access to enterprise applications are highly valued. Hijacked session tokens from services like Microsoft 365 or Google Workspace can expose internal emails, cloud storage, and critical business applications. Azure Entra ID tokens offer significant appeal due to their capability to grant long-term access without reauthentication, even across browser sessions.

Within a compromised environment, attackers can manipulate existing enterprise applications, using tools like TokenSmith, ROADtools, and AADInternals. These utilities allow for the manipulation of tokens, escalating access, or pivoting across services within the tenant. Although Conditional Access Policies (CAPs) can block unauthorized login attempts based on location or device compliance, they are not foolproof. Varonis demonstrated that attackers could emulate a victim’s typical environment to bypass CAP restrictions.

Mitigation Strategies and the Future of Attack Prevention

Researchers at Varonis Threat Labs have discovered an intricate method that cyber attackers use to bypass multi-factor authentication (MFA) and gain persistent access to enterprise cloud environments. This method involves the exploitation of browser cookies, allowing attackers to completely evade conventional security defenses and remain unnoticed. The research, which was unveiled in April, sheds light on the increasing sophistication of session hijacking techniques. These techniques involve the use of browser extensions and scripts to steal authentication tokens, enabling cybercriminals to maintain prolonged access to secure systems. The findings emphasize a concerning progression in cyber attack strategies, suggesting that even advanced security measures like MFA might not be sufficient to protect against these crafty techniques. As cyber threats continue to evolve, it becomes evident that organizations must continually adapt their security protocols to stay ahead of potential breaches. This research serves as a crucial reminder of the need for constant vigilance and reassessment of security frameworks to protect sensitive data in enterprise cloud environments.

Explore more

BSP Boosts Efficiency with AI-Powered Reconciliation System

In an era where precision and efficiency are vital in the banking sector, BSP has taken a significant stride by partnering with SmartStream Technologies to deploy an AI-powered reconciliation automation system. This strategic implementation serves as a cornerstone in BSP’s digital transformation journey, targeting optimized operational workflows, reducing human errors, and fostering overall customer satisfaction. The AI-driven system primarily automates

Is Gen Z Leading AI Adoption in Today’s Workplace?

As artificial intelligence continues to redefine modern workspaces, understanding its adoption across generations becomes increasingly crucial. A recent survey sheds light on how Generation Z employees are reshaping perceptions and practices related to AI tools in the workplace. Evidently, a significant portion of Gen Z feels that leaders undervalue AI’s transformative potential. Throughout varied work environments, there’s a belief that

Can AI Trust Pledge Shape Future of Ethical Innovation?

Is artificial intelligence advancing faster than society’s ability to regulate it? Amid rapid technological evolution, AI use around the globe has surged by over 60% within recent months alone, pushing crucial ethical boundaries. But can an AI Trustworthy Pledge foster ethical decisions that align with technology’s pace? Why This Pledge Matters Unchecked AI development presents substantial challenges, with risks to

Data Integration Technology – Review

In a rapidly progressing technological landscape where organizations handle ever-increasing data volumes, integrating this data effectively becomes crucial. Enterprises strive for a unified and efficient data ecosystem to facilitate smoother operations and informed decision-making. This review focuses on the technology driving data integration across businesses, exploring its key features, trends, applications, and future outlook. Overview of Data Integration Technology Data

Navigating SEO Changes in the Age of Large Language Models

As the digital landscape continues to evolve, the intersection of Large Language Models (LLMs) and Search Engine Optimization (SEO) is becoming increasingly significant. Businesses and SEO professionals face new challenges as LLMs begin to redefine how online content is managed and discovered. These models, which leverage vast amounts of data to generate context-rich responses, are transforming traditional search engines. They