New release · Cybersecurity & AI

Combating Cyberattacks
Targeting the AI Ecosystem

Second Edition

A deep technical exploration of the AI-driven cyber threat landscape — covering LLMs, Generative AI, agentic systems, prompt injection, and the defensive frameworks to stop them.

Get the book → What's inside
De Gruyter · 2nd Ed.
Combating Cyberattacks
Targeting the
AI Ecosystem
Aditya K Sood
Coverage

The modern AI attack surface

⟦LLM⟧
Large Language Models
Inherent risks in LLM architectures, model behavior vulnerabilities, and exploitation techniques targeting the inference layer.
∅↯
Prompt Injection
Direct and indirect injection attacks, jailbreak methodologies, and prompt engineering as a vector for system compromise.
⟳⚡
Agentic Systems
Autonomous agent manipulation, tool abuse, multi-model orchestration risks, and agent chaining vulnerabilities.
▩▩▩
GenAI Applications
Generative AI application security, RAG pipeline risks, data leakage through model outputs, and API-level attack surfaces.
⌬◈
AI Infrastructure
System-level integrations, training pipeline integrity, model supply chain risks, and infrastructure hardening.
⛨→
Defensive Frameworks
Structured mitigation strategies, secure design patterns, and a comprehensive framework for defending modern AI systems.
What you'll learn

From adversarial thinking
to robust defence

01 Identify and assess vulnerabilities across the full AI pipeline — from model behaviour through to system-level integrations — using real-world case studies.
02 Understand how prompt engineering, autonomous agents, and multi-model orchestration introduce novel attack surfaces.
03 Apply hands-on technical methodologies for adversarial testing of LLMs, GenAI applications, and agentic architectures.
04 Implement a structured security framework: from foundational AI concepts through practical assessment to robust mitigation controls.
05 Defend against evolving AI-powered threats in increasingly autonomous and interconnected environments.
Who this is for
Security practitioners IT professionals AI/ML engineers Business leaders Red teamers Researchers
2nd
Edition
ISBN
9781501525285
DeG
De Gruyter
Author
AKS

Aditya K Sood

Cybersecurity Researcher & Author

Aditya K Sood is a seasoned cybersecurity researcher and practitioner whose work bridges the gap between academic rigour and real-world adversarial thinking. With deep expertise in threat intelligence, exploit research, and secure systems design, he has spent years analysing how emerging technologies — particularly AI — expand and reshape the attack surface for defenders.

Website Publisher page
Areas of expertise
AI security & adversarial machine learning
LLM vulnerability assessment and red-teaming
Threat intelligence and exploit research
Secure design of agentic and GenAI systems
Malware analysis and infrastructure security
Now available

Ready to defend the AI frontier?

Whether you're securing an enterprise AI stack or red-teaming a GenAI application, this book equips you with the knowledge and techniques to stay ahead of the threat.

Order from De Gruyter → More by this author