Back
#268
March 23, 2026

EP268 Weaponizing the Administrative Fabric: Cloud Identity and SaaS Compromise in M Trends 2026

Guests:

Topics:

Threat Intelligence
29:29

Subscribe at YouTube

Subscribe at Spotify

Subscribe at Apple Podcasts

Topics covered:

  • Do we need to rethink "Mean Time to Respond" entirely, or are we just in deep trouble?
  • Why are threat groups collaborating so well, and are there actual lessons for defenders in their "business" model?
  • What is the scalable advice for teams worried about voice phishing and GenAI cloning?
  • What does "weaponizing the administrative fabric" actually mean in a world where identity is the perimeter?
  • Why is identity/SaaS compromise "news" in 2026 when cloud security folks have been shouting about it for years? What actually changed?
  • What’s the latest in supply chain compromise, particularly regarding malicious open-source packages?
  • How do we defend against malware that is "lazy" enough to use the victim’s own AI tools for reconnaissance?
  • What is the specific advice for Detection and Response (D&R) teams to handle "living off the land" (or "living off the cloud")?
  • How do you fix the situation when IT and Security departments genuinely hate each other?
  • Besides reading the report, what is the one book or piece of advice for a CISO to survive this year?

Do you have something cool to share? Some questions? Let us know:

Transcript

The discussion centered on the release of the 2026 M-Trends report, Mandiant’s flagship publication detailing the current state of global cyber threats based on frontline incident response data. The overarching theme of this year’s findings is the staggering acceleration of attacker efficiency, driven by automated collaboration between disparate threat groups.

The conversation pivoted from technical metrics, such as the collapse of "handoff time" from hours to seconds, to the psychological and organizational challenges facing defenders. A key takeaway is that "getting good at cyber" is no longer purely a technical pursuit; it requires a fundamental shift toward better internal organizational policy, leadership alignment, and the dismantling of antagonistic relationships between IT and security teams.

Detailed Discussion Points

The 22-Second Handoff: The New Speed of Compromise

The most alarming statistic highlighted is the reduction in "handoff time"—the duration between an initial access provider gaining entry and a secondary actor (often a ransomware operator) receiving control. This metric has plummeted from approximately eight hours to a mere 22 seconds.

Automation as the Catalyst: This speed is not achieved by human manual labor but through highly automated tooling. When Group A breaches a system, they immediately deploy Group B’s toolset via automated scripts.

Defensive Implications: Traditional Mean Time to Respond (MTTR) metrics are becoming obsolete if they cannot account for sub-minute escalations. Defenders can no longer afford to "bucket" low-impact indicators (like maladvertising) for later review, as these now serve as the immediate precursor to high-impact breaches.

The Weaponization of the Administrative Fabric

The panel discussed the shift from "breaking in" to "logging in." Attackers are increasingly targeting the "administrative fabric"—the identity providers and SaaS integrations that bypass the traditional network perimeter.

Identity as the Perimeter: Rather than attacking domain controllers directly, adversaries are compromising identity infrastructure to "spider" out into third-party integrations.

Living off the Cloud: Much of this activity involves no malware. Attackers utilize legitimate credentials to navigate environments, making detection exceptionally difficult without a deep understanding of baseline user behavior.

Supply Chain and AI-Enhanced Malware

The report notes a persistent trend in supply chain compromises, specifically through Trojanized open-source packages.

AI-Enabled Reconnaissance: In a notable 2026 case study, the "QuietVault" malware was observed searching for AI command-line interfaces on infected systems. Upon finding one, the malware issued prompts to the local AI to perform reconnaissance and data collection on behalf of the attacker—effectively using the victim's own tools to automate the breach.

The Crisis of Leadership and Collaboration

The guests argued that many security failures are rooted in poor organizational culture.

The IT-Security Divide: In organizations where IT and security are antagonistic, security posture inevitably suffers. The solution is not better technology, but better leadership from the executive level to ensure all departments are "pulling in the same direction."

Trust but Verify: While outsourcing (HR, Help Desk, etc.) is a business necessity, it introduces critical risks. Organizations must move beyond contractual trust to implementing hard controls and monitoring for these third-party relationships.

Topic Timeline

Introduction and the New Format: Introduction of hosts and guests; discussion of the shift toward video and YouTube Shorts for 2026.

The Philosophy of Incident-Based Learning: A debate on whether security can be learned solely from incidents versus first principles (using the "plane crash" and physics analogies).

M-Trends 2026 Metric Shock: Deep dive into the reduction of attacker handoff time from eight hours to 22 seconds.

The Mechanics of Group Collaboration: How automated partnerships between initial access brokers and ransomware groups function.

Defensive Pressure and MTTR: The need for defenders to stop treating "low impact" alerts as low priority given the speed of escalation.

Initial Infection Vectors: Discussion on maladvertising and "click-fix" lures as the primary gates for modern breaches.

The Evolution of Voice Phishing (Vishing): How generative AI and voice cloning have made social engineering more convincing and scalable.

Weaponizing the Administrative Fabric: Analysis of how identity and SaaS integrations have become the primary targets for lateral movement.

The "Logging In" Reality: Addressing the trend of attackers using legitimate credentials rather than exploiting software vulnerabilities.

Supply Chain Threats and North Korean Tactics: Discussion of Trojanized dependencies in open-source registries and social engineering of developers.

AI-Driven Malware: A case study of malware using a system's own AI CLI to perform reconnaissance.

The "Be Better at Policy" Argument: Shifting the focus from technical excellence to soft skills, collaboration, and executive leadership.

Dismantling Antagonistic Relationships: Strategies for fixing the culture between IT and security teams.

Third-Party and Outsourced Risk: The importance of monitoring help desks and outsourced HR functions.

Closing Recommendations: Guidance for CISOs on using the report; recommended reading (Axelrod’s Evolution of Cooperation).

View more episodes