Cyber Security
+ AI Agents

Concepts: What is Cybersecurity? Why it matters more than ever CIA Triad — Confidentiality, Integrity, Availability Types of Hackers — White, Grey, Black Hat Evolution of Cyber Threats — from script kiddies to AI-powered attacks The convergence of AI and Cybersecurity — 2026 threat landscape Career paths in AI-era cybersecurity India's regulatory landscape — DPDP Act, CERT-In guidelines, RBI/SEBI mandates Hands-On Lab: Setting up your security lab — VirtualBox / VMware Installing Kali Linux, Parrot OS Lab environment walkthrough and safety protocols

Concepts: Windows vs Linux security fundamentals Linux file system architecture Essential Linux commands for security professionals File permissions, ownership, and access control Process management and system monitoring Bash scripting fundamentals for automation Hands-On Lab: Linux terminal mastery exercises File permission and privilege configuration lab Writing basic Bash scripts for system auditing Windows Event Viewer and Sysmon introduction

Concepts: Network fundamentals — what every security professional must know OSI Model and TCP/IP stack — deep dive IP addressing, subnetting, MAC addresses, ports TCP vs UDP — protocol behavior and security implications Core protocols — HTTP/S, FTP, DNS, SSH, SMTP, DHCP Network devices — routers, switches, firewalls Network segmentation and micro-segmentation concepts Hands-On Lab: Network commands — ifconfig, ip, ping, traceroute, netstat, ss Wireshark packet capture and analysis (basics) Understanding DNS resolution and HTTP traffic Network mapping with basic Nmap scans

Concepts: Firewalls — types, rules, configuration principles Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS) VPN fundamentals and secure tunneling Zero Trust Architecture (ZTA) — principles and why it matters in 2026 Common network attacks — MITM, ARP Spoofing, DNS Poisoning, DDoS Network monitoring and traffic analysis Hands-On Lab: Wireshark deep traffic analysis Simulated MITM demonstration (controlled lab environment) Firewall rule configuration exercise Introduction to Snort/Suricata for IDS

Concepts: What is Ethical Hacking? Penetration Testing lifecycle Legal and ethical aspects — scope, authorization, responsible disclosure Bug bounty programs — how they work, major platforms Reconnaissance — passive vs active techniques OSINT (Open Source Intelligence) fundamentals Google Dorking, WHOIS, DNS enumeration Tools & Hands-On Lab: Setting up vulnerable targets — DVWA, Metasploitable, Juice Shop Nmap scanning and service discovery theHarvester, WHOIS lookups Shodan exploration (demo/walkthrough)

Concepts: Port scanning techniques and strategies Service and version enumeration Vulnerability scanning methodology CVSS scoring and vulnerability prioritization Predictive vulnerability management — how AI is changing this Tools & Hands-On Lab: Advanced Nmap techniques — scripts, OS detection, service enumeration Nikto web server scanning OpenVAS / Nessus introduction Scanning vulnerable VMs and analyzing results Writing a basic vulnerability assessment report

Concepts: How modern web applications work — client-server architecture, APIs OWASP Top 10 (Web Applications) — comprehensive walkthrough SQL Injection — types, detection, exploitation, prevention Cross-Site Scripting (XSS) — stored, reflected, DOM-based Cross-Site Request Forgery (CSRF) File upload vulnerabilities, IDOR, broken authentication API security fundamentals Tools & Hands-On Lab: SQL Injection lab on DVWA XSS exploitation and mitigation lab Burp Suite introduction — intercepting, modifying, replaying requests OWASP ZAP for automated scanning

Concepts: Password attack methodologies — brute force, dictionary, credential stuffing Hashing algorithms, salting, password storage best practices Privilege escalation — basic techniques (Linux and Windows) Malware types — viruses, trojans, ransomware, worms, rootkits Ransomware-as-a-Service (RaaS) — the 2026 threat Introduction to wireless security — WPA2/WPA3, Evil Twin (theory) Tools & Hands-On Lab: Hydra — brute force attacks John the Ripper and Hashcat — password cracking Basic privilege escalation exercises Malware analysis concepts (static analysis introduction)

Concepts: What is Artificial Intelligence? — types and capabilities Machine Learning fundamentals — supervised, unsupervised, reinforcement learning Training vs inference — the ML lifecycle Neural networks and deep learning (conceptual overview) How AI is transforming both cyberattacks and cyberdefense Python for security — why every security professional needs it Hands-On Lab: Python environment setup — Jupyter Notebook, VS Code Python essentials for security — scripting, file handling, API calls Simple ML model demonstration using Scikit-learn Dataset loading, exploration, and visualization with Pandas/Matplotlib

Concepts: What are Large Language Models (LLMs)? How they work Generative AI — capabilities and limitations AI Agents — what they are, how they differ from chatbots The Agentic AI revolution — autonomous planning, tool use, decision-making AI agent architectures — single agent, multi-agent, orchestration The expanded attack surface of AI systems — data, model, API, infrastructure Why AI agent security is fundamentally different from traditional LLM security Hands-On Lab: Interacting with LLM APIs (OpenAI, Anthropic, Google) Understanding token limits, system prompts, and model behavior Mapping the attack surface of a sample AI application AI threat modeling exercise using STRIDE methodology

Concepts: AI-enhanced phishing and social engineering — deepfakes, voice cloning AI-generated malware and polymorphic threats Automated reconnaissance and AI-powered vulnerability discovery Cybercrime-as-a-Service (CaaS) — AI-powered underground tools AI-assisted password cracking and credential stuffing Autonomous attack agents — what they can do today Case studies — real-world AI-powered cyberattacks (2024–2026) Hands-On Lab: Analyzing AI-generated phishing emails — detection techniques Deepfake detection demonstration Understanding AI-augmented attack workflows Threat intelligence gathering using AI tools

Concepts: Adversarial Machine Learning — core concepts Adversarial examples — image perturbation, text manipulation White-box vs black-box attacks on ML models Data poisoning attacks — clean-label poisoning, backdoor attacks Training-time vs inference-time attacks Data leakage risks in AI pipelines Bias injection and fairness manipulation Hands-On Lab: Creating adversarial image examples — model misclassification demo Data poisoning simulation — observing accuracy degradation Dataset inspection for anomalies and poisoned data Bias detection in training datasets

Concepts: OWASP Top 10 for LLM Applications (2025 Edition) — complete walkthrough LLM01: Prompt Injection (Direct & Indirect) LLM02: Sensitive Information Disclosure LLM03: Supply Chain Vulnerabilities LLM04: Data and Model Poisoning LLM05: Improper Output Handling LLM06: Excessive Agency LLM07: System Prompt Leakage LLM08: Vector and Embedding Weaknesses LLM09: Misinformation LLM10: Unbounded Consumption Prompt injection deep dive — techniques, real-world examples, defenses Jailbreaking LLMs — methods and countermeasures Hallucinations as a security risk Hands-On Lab: Prompt injection attack exercises — direct and indirect Jailbreak attempt analysis — safe vs unsafe prompts Red-teaming LLM responses — systematic approach Implementing basic prompt guardrails and input validation

Concepts: Model extraction attacks — stealing model weights and behavior Model inversion attacks — recovering training data Membership inference attacks — determining if data was in training set AI intellectual property theft and protection AI supply chain risks — compromised models, poisoned datasets, malicious packages Shadow AI — unmonitored AI tools within organizations Secure model deployment pipelines Hands-On Lab: API abuse simulation — querying models to extract behavior Model behavior observation and fingerprinting Scanning for vulnerable AI/ML dependencies Secure model deployment checklist exercise

Concepts: Why agentic AI requires a completely new security framework OWASP Top 10 for Agentic Applications — comprehensive walkthrough: ASI01: Agent Goal Hijacking ASI02: Tool Misuse & Unintended Actions ASI03: Insecure Agent-to-Agent Communication ASI04: Insufficient Agent Authorization ASI05: Sensitive Data Leakage ASI06: Knowledge Base Poisoning ASI07: Denial of Wallet / Unbounded Resource Consumption ASI08: Rogue Agents & Cascading Failures ASI09: Inadequate Audit & Observability ASI10: Insecure Agent Memory & Context Principle of Least Agency — the foundational defense principle Agent identity management — non-human identity security Multi-agent security patterns Hands-On Lab: Agent goal hijacking simulation Tool misuse scenario analysis Designing secure agent authorization frameworks Agent audit logging and observability exercise

Concepts: Secure AI/ML pipeline design — from data to deployment Model hardening techniques — adversarial training, input validation AI model monitoring in production — drift detection, behavioral anomalies Output filtering and Data Loss Prevention (DLP) for AI AI governance frameworks overview: NIST AI Risk Management Framework (AI RMF) EU AI Act — high-risk AI requirements (enforcement August 2026) ISO/IEC 42001 — AI Management Systems India's DPDP Act implications for AI Responsible AI — explainability (XAI), fairness, accountability AI red teaming methodology Hands-On Lab: Improving model robustness — adversarial training demo Explainability demonstration using SHAP/LIME AI risk assessment worksheet exercise Building an AI governance checklist for an enterprise

Concepts: The modern Security Operations Center (SOC) — AI-enhanced operations SIEM fundamentals — log aggregation, correlation, alerting AI-powered threat detection — behavioral analytics, anomaly detection Predictive threat modeling using AI Automated incident triage and response AI-powered vulnerability prioritization Threat intelligence platforms and AI-driven threat hunting Reducing alert fatigue with AI — intelligent alert correlation Tools & Hands-On Lab: Splunk fundamentals — log ingestion, search, dashboards, alerting AI-assisted log analysis exercise Creating detection rules based on behavioral patterns Windows Event Viewer deep dive and Sysmon configuration Threat hunting scenario walkthrough

Concepts: Cloud security fundamentals — shared responsibility model AWS / Azure / GCP security services overview Identity and Access Management (IAM) in cloud environments Cloud-native security architectures — continuous authentication and monitoring Securing AI workloads in the cloud (AWS Bedrock, Azure AI, GCP Vertex) Containerization security (Docker, Kubernetes basics) Infrastructure as Code (IaC) security scanning Cloud misconfiguration — the #1 cloud vulnerability Hands-On Lab: Cloud security configuration review exercise IAM policy analysis and least privilege exercise Cloud security audit simulation Securing an AI deployment in cloud environment

Concepts: Incident response lifecycle — preparation, detection, containment, eradication, recovery, lessons learned AI-powered incident response — automated containment and triage Digital forensics introduction — evidence collection, chain of custody AI-enhanced forensics — automated log correlation, timeline reconstruction Incident response for AI system failures — unique considerations AI supply chain incident management Business continuity and disaster recovery in AI-era Compliance incident reporting — CERT-In, DPDP Act requirements Hands-On Lab: Incident response tabletop exercise AI incident scenario analysis — compromised agent response Log forensics using Splunk Creating an incident response playbook for AI systems

Concepts: Post-Quantum Cryptography (PQC) — why it matters now Quantum computing threats to current encryption NIST PQC standards and migration planning Zero Trust Architecture (ZTA) — deep dive implementation Identity security in the agentic era — machine identities, non-human identities Passwordless authentication, continuous verification Supply chain security — software and AI supply chains Cybersecurity mesh architecture (CSMA) Regulatory landscape 2026+ — EU AI Act, Colorado AI Act, India DPDP Hands-On Lab: Zero Trust policy design exercise Identity and access management review Quantum-safe encryption concepts demonstration Supply chain security assessment exercise

Concepts: Red Team operations — planning, execution, reporting Blue Team operations — detection, response, hardening Purple Team — collaborative security improvement AI Red Teaming — testing AI systems for vulnerabilities MITRE ATT&CK framework — tactics, techniques, procedures MITRE ATLAS — adversarial threat landscape for AI Building detection rules and hunting hypotheses Writing professional penetration test reports Hands-On Lab: Red Team exercise — full attack chain on practice target Blue Team exercise — detecting and responding to the attack AI red teaming — testing an LLM application for OWASP vulnerabilities MITRE ATT&CK mapping exercise

Concepts: CTF methodology and competitive cybersecurity Challenge categories — web, forensics, crypto, reverse engineering, AI TryHackMe and Hack The Box guided exercises AI security-specific CTF challenges Hands-On Lab: Full CTF competition TryHackMe / Hack The Box challenge labs AI security challenge — find and exploit LLM vulnerabilities Team-based red/blue team simulation

Students choose one major project and present it: Project Option A: Enterprise Penetration Test + AI Security Assessment Project Option B: Secure AI Agent Pipeline Design Project Option C: AI-Powered SOC — Detection & Response Project Option D: AI Governance Framework for an Indian Enterprise

Certification Roadmap Guidance: Foundation: CompTIA Security+, CEH (EC-Council), eJPT (INE) Intermediate: OSCP, BTL1 (Blue Team Level 1), CompTIA CySA+ AI Security: CAISP (Certified AI Security Professional), NVIDIA Agentic AI Governance: AIGP (IAPP), ISO 42001 Practitioner Advanced: CISSP, OSWE, GIAC certifications Career Preparation: Resume building for AI-cybersecurity roles — what recruiters want in 2026 LinkedIn profile optimization and personal branding Portfolio building — GitHub, blog posts, CTF writeups Interview preparation — technical + behavioral Mock interviews with industry feedback Job search strategies — direct applications, bug bounties, freelancing, consulting