Traversing the Cyber Threat Kill Chain with Composite AI

Blog
Tue Dec 10 2024

Traversing the Cyber Threat Kill Chain with Composite AI

Aman Shrivastava
Threat Hunter III, AiStrike
Modern cybersecurity teams are overwhelmed by a relentless flood of alerts, each demanding attention. Investigating these in isolation not only drains analysts but also obscures the bigger picture—how individual alerts might connect to a broader, more sophisticated attack.
Table of Contents

Modern cybersecurity teams are overwhelmed by a relentless flood of alerts, each demanding attention. Investigating these in isolation not only drains analysts but also obscures the bigger picture—how individual alerts might connect to a broader, more sophisticated attack.

Threat actors like Black Basta exploit this lack of context, deploying advanced tactics that span multiple stages of the kill chain. To counter these threats, organizations need a smarter approach: one that prioritizes alerts and maps their progression across the kill chain to uncover relationships and true intent.

This blog explores how AiStrike leverages Composite AI and the MITRE ATT&CK framework to contextualize alerts, reduce fatigue, and empower analysts to stay ahead of complex threats like Black Basta.


Black Basta - Introduction

Black Basta, a ransomware group first identified in 2022, is experiencing a resurgence in 2024, continuing to pose significant threats globally. The group operates under a Ransomware-as-a-Service (RaaS) model, leveraging sophisticated tactics such as double extortion—encrypting data and threatening to leak sensitive information if the ransom isn't paid.

Recent activity demonstrates Black Basta's use of advanced tools and techniques, including exploiting vulnerabilities like CVE-2024-1709 and deploying tools such as Mimikatz and Cobalt Strike for privilege escalation and lateral movement. They also utilize spear-phishing campaigns and compromised remote access tools like AnyDesk to gain initial access. After infiltration, they disable defenses using tools like "Backstab" and encrypt systems with ChaCha20 and RSA-4096 algorithms. Indicators suggest increased operational activity in late 2024, with new infrastructure supporting their campaigns​

The group’s resurgence underscores the importance of enhanced cybersecurity measures and vigilance against ransomware threats. More detailed indicators and mitigation strategies are available in the CISA advisory


Key Attributes and Threat Profile:

1. Geographic Focus and Targeting:

  • Targeted countries: North America, Europe, Japan, Australia, and New Zealand.
  • Targeted sectors: critical infrastructure, healthcare, and finance.

2. Threat Actors Aliases:

  • Conti Ransomware: Black Basta is believed to have ties to Conti, evident from shared tactics, infrastructure, and code similarities.
  • FIN7 (Carbanak): The group shares custom EDR evasion modules and command-and-control (C2) methods with FIN7.

3. Tools used:

BITSAdmin,Mimikatz,RClone,Splashtop,WMI,CobaltStrike,SoftPerfect,WinSCP,Mimikatz,Qakbot,

PowerShell,ScreenConnect,PSExec,EvilProxy,SystemBC,Backstab,Netcat,Quick Assist,NetSupport Manager

4. CVE Exploited:

CVE-2024-1709,CVE-2024-26169,CVE-2020-1472,CVE-2021-42278,CVE-2021-42287,CVE-2021-34527

Attack Lifecycle and Techniques:

Ref: https://www.cisa.gov/news-events/cybersecurity-advisories/aa24-131a

1. Initial Access:

2. Execution:

3. Privilege Escalation:

4. Defence Evasion

5.Impact:

Mitre Based Grouping to detect Black basta :

In this scenario, our AI Agents are trained to successfully correlate multiple low-priority alerts into a high-risk threat, aligning with techniques outlined in the MITRE ATT&CK framework and indicating the presence of Black Basta ransomware activity.

The system effectively linked vulnerability data to our emerging threats database, which has been generated using LLM based AI agents, trained on over 10 years of threat campaign data. This AI agent leveraged the NER(Named entity recognition) to read and understand the details of the campaign, associating it with relevant alerts.

Sequence of Events :

Black Basta Context:

This pattern strongly aligns with Black Basta ransomware campaigns, where attackers leverage phishing for initial access, execute malicious payloads to establish C2, and deploy ransomware after disabling recovery options. The group’s use of sophisticated EDR evasion techniques and MITRE ATT&CK-aligned behaviors makes it a significant threat.

Additionally, AiStrike’s AI Copilot guides analysts through the investigation process, highlighting critical threat details and recommending pre-configured remediation actions built into the platform. Analysts can interact with the Copilot using prompts tailored to each alert type.

Conclusion:

At AiStrike, we harness the power of Composite AI to transform cybersecurity, enhancing defenses at every level. Our platform seamlessly integrates with enterprise security tools like SIEM, CNAPP, and XDR, ensuring compatibility and maximizing the value of existing solutions.

With AI-driven alert classification and automated investigations, AiStrike enables analysts to correlate events across diverse systems, uncover hidden patterns, and visualize complex relationships. This streamlines incident investigations, significantly reducing the time and effort required for effective threat response.

Our dynamic prioritization system, powered by an AI agent and machine learning classification engine, automatically correlates seemingly unrelated alerts as they progress through the kill chain. It performs behavioral analytics and analyzes network traffic patterns, such as robotic activity, impossible geotravel, rare events, or activity spikes, identifying common features and entities to deliver actionable insights.

By embracing Composite AI, organizations achieve unparalleled efficiency and precision in threat detection and mitigation, staying ahead of attackers. This innovative approach ensures smarter, more resilient defenses against the ever-evolving cyber threat landscape.

Latest Resources

All Resources
Blog

Investigating millions of CSPM alerts — where do you even start?

I got this question last week from one of the largest financial institutions: “When you’re looking at millions of CSPM alerts, do you actually investigate them or just treat them as hygiene issues and assign them to the cloud team?” Honestly, it’s a fair question—and one a lot of teams are probably asking themselves.
Read More
Blog

Rethinking Alert Ownership in Security Ops

All alerts are not equal. Yet somehow, every alert becomes the SOC’s problem. Every day, SIEM and CNAPP tools flood the SOC with alerts — but take a closer look, and they generally fall into four categories:
Read More
Blog

Blind Spots vs. False Positives — Which One Kills Faster?

Most SOCs worry about false positives — the noisy alerts that eat away analyst time and slow down response. But what if the real killer isn’t the noise you hear, but the silence you don’t?
Read More
Case study

How Sunrun Transformed Security Operations with AiStrike

Transforming to an AI-Powered Self-Improving SOC
Read More
Case study

Global Software Design Company Leverages AiStrike to Investigate Cloud Alerts

Global Software Design Company Leverages AiStrike to Investigate Cloud Alerts
Read More
News

Harsh Patwardhan Joins AiStrike as Chief Technology Officer

Reuniting a Proven Leadership Team to Build the Future of Autonomous Security Operations.
Read More
News

AiStrike Announces AI Agents for Detection Optimization, Advancing the Complete AI-Augmented SOC

San Francisco, CA – April 14, 2025 – AiStrike, the AI SOC automation platform transforming cybersecurity operations, today announced the launch of its AI Agents for Detection Optimization—a first-of-its-kind capability that helps security teams improve detection quality, eliminate blind spots, and reduce alert noise by automatically identifying coverage gaps and tuning detections in real time.
Read More
News

AiStrike Emerges from Stealth to Solve Cloud Security Investigation and Response using AI-powered Automation

Guidelines for selecting the most suitable CMS for your project.
Read More
News

Cloud Security Operations Leader AiStrike Launches AI-Powered Cloud Security Investigation and Response Solution on AWS Marketplace

Exploring the advantages of utilizing a CMS for website management.
Read More
News

Jhilmil Kochar Joins AiStrike as Chief Engineering and Product Leader

Former CrowdStrike Executive with over 30 years of experience in Cybersecurity and Product Development joins AiStrike, the startup redefining AI-Powered Security Automation.
Read More
Datasheets

AI-Powered Automation for Threat Investigation and Response

In today's landscape of relentless cyber-attacks, organizations are facing increasing threats to their critical assets. Security detection tools like SIEM, XDR, and CNAPP generate vast volumes of alerts—often lacking sufficient context—leaving security teams overwhelmed with alert backlog. With limited resources and insufficient business context, prioritizing critical alerts that require immediate action becomes a significant challenge.
Read More
Solution Briefs

AiStrike for AWS

Cloud infrastructure today is the primary target for malicious actors. The risk of exposure of cloud assets continues to grow as organizations expand their cloud footprint and new cyberattacks targeting cloud infrastructure emerge.
Read More
White Papers

CISO Guide: AI-Automated Cloud Security Operations

This guide provides CISOs with a comprehensive understanding of how AI-driven automation can revolutionize cloud security operations, enhancing both efficiency and effectiveness.
Read More
Blog

Investigating millions of CSPM alerts — where do you even start?

I got this question last week from one of the largest financial institutions: “When you’re looking at millions of CSPM alerts, do you actually investigate them or just treat them as hygiene issues and assign them to the cloud team?” Honestly, it’s a fair question—and one a lot of teams are probably asking themselves.
Read More
Blog

Rethinking Alert Ownership in Security Ops

All alerts are not equal. Yet somehow, every alert becomes the SOC’s problem. Every day, SIEM and CNAPP tools flood the SOC with alerts — but take a closer look, and they generally fall into four categories:
Read More
Blog

Blind Spots vs. False Positives — Which One Kills Faster?

Most SOCs worry about false positives — the noisy alerts that eat away analyst time and slow down response. But what if the real killer isn’t the noise you hear, but the silence you don’t?
Read More