AI Cybersecurity Course

Lesson 1 • Core Cybersecurity Concepts

Covers the CIA triad, threat actors, and attack surfaces. Provides the vocabulary needed for all subsequent AI-security integration topics.

Lesson 2 • AI and Cybersecurity Convergence

Examines why AI is adopted in security operations and where it adds measurable value. Sets expectations for the course's applied focus.

Lesson 3 • Threat Landscape and Data Ecosystems

Maps modern threat categories to the data environments AI systems must protect. Grounds later detection techniques in real-world attack patterns.

Lesson 1 • Security Data Sources and Types

Identifies logs, network flows, endpoint telemetry, and threat feeds as primary inputs. Explains how data type affects model selection.

Lesson 2 • Data Labelling and Annotation

Teaches manual and semi-automated labelling of security events for supervised learning. Addresses label quality's direct impact on model accuracy.

Lesson 3 • Data Quality and Preprocessing

Addresses noise reduction, normalisation, and feature extraction from raw telemetry. Prepares learners to deliver model-ready datasets.

Lesson 1 • Deep Learning in Threat Detection

Applies CNNs and RNNs to malware classification and sequential log analysis. Highlights trade-offs between model complexity and operational speed.

Lesson 2 • Handling Class Imbalance in Security Data

Addresses the rarity of attack samples relative to benign traffic using resampling and cost-sensitive learning. Improves minority-class detection rates.

Lesson 3 • Unsupervised Anomaly Detection

Uses clustering and density-based methods to surface unknown threats without labels. Demonstrates value for zero-day and insider threat scenarios.

Lesson 4 • Supervised Detection Models

Trains classifiers such as decision trees, random forests, and gradient boosting on labelled attack data. Connects algorithm choice to detection objectives.

Lesson 1 • Threat Intelligence Text Mining

Extracts indicators of compromise and tactics from unstructured threat reports. Automates intelligence enrichment for faster analyst response.

Lesson 2 • NLP Fundamentals for Security Text

Covers tokenisation, embeddings, and language models applied to security corpora. Establishes the text-processing foundation for all NLP security tasks.

Lesson 3 • Phishing and Social Engineering Detection

Trains classifiers to identify phishing emails, malicious URLs, and deceptive messages. Directly addresses one of the highest-volume attack vectors.

Lesson 1 • SOC Performance Measurement

Defines KPIs for AI-augmented SOC operations including mean time to detect and respond. Enables continuous improvement of AI-human workflows.

Lesson 2 • Threat Hunting with AI Assistance

Combines analyst hypothesis generation with AI-driven data exploration to find hidden threats. Builds proactive hunting skills beyond reactive alerting.

Lesson 3 • AI-Augmented Alert Triage

Applies machine learning scoring to prioritise alerts by severity and confidence. Reduces analyst workload while maintaining detection coverage.

Lesson 4 • Security Orchestration and AI Integration

Connects AI models to SOAR platforms for automated playbook execution. Demonstrates end-to-end automation from detection to containment.

Lesson 1 • Data Poisoning and Supply Chain Attacks

Examines how corrupted training data degrades model integrity and covers detection and mitigation strategies. Links to AI supply chain risk.

Lesson 2 • Evasion Techniques and Defences

Demonstrates how adversaries craft inputs to bypass machine learning-based detectors and teaches adversarial training as a countermeasure.

Lesson 3 • Adversarial Machine Learning Fundamentals

Defines evasion, poisoning, and model inversion attacks against security AI. Establishes the adversarial threat model for all subsequent defences.

Lesson 1 • Risk-Based Vulnerability Prioritisation

Applies machine learning scoring to rank vulnerabilities by exploitability, asset criticality, and threat context. Focuses remediation effort on highest-impact issues.

Lesson 2 • Patch and Remediation Optimisation

Uses AI to sequence patching actions, predict regression risk, and validate remediation effectiveness. Closes the vulnerability management lifecycle.

Lesson 3 • AI-Driven Vulnerability Discovery

Uses fuzzing, static analysis, and machine learning models to surface software vulnerabilities faster than manual review. Integrates with existing scanning toolchains.

Lesson 1 • Regulatory and Compliance Alignment

Aligns AI security deployments with data protection, privacy, and sector-specific compliance obligations. Avoids jurisdiction-specific codes; focuses on functional requirements.

Lesson 2 • Bias, Fairness, and Explainability

Addresses demographic bias in security models and applies explainability methods to support analyst trust. Connects fairness to operational and legal risk.

Lesson 3 • Strategic AI Security Roadmap

Guides learners through building a multi-year AI security adoption plan tied to business objectives and maturity levels. Produces a board-ready investment narrative.

Lesson 4 • Continuous Model Monitoring and Governance

Establishes processes for detecting model drift, retraining triggers, and ongoing performance audits in production. Sustains AI security effectiveness over time.