Moderation Course

Lesson 1 • Types of Harmful Content

Catalogues major harm categories including hate speech, harassment, misinformation, and illegal content. Provides a taxonomy used throughout the course for classification tasks.

Lesson 2 • Core Ethical Principles

Introduces proportionality, consistency, communication transparency, and due process as guiding values. Frames ethical reasoning as a practical tool moderators apply in every decision.

Lesson 3 • Stakeholders and Platform Ecosystems

Identifies users, platforms, advertisers, and regulators as interdependent actors. Explains how stakeholder interests shape moderation priorities and decisions.

Lesson 4 • What Content Moderation Is

Defines moderation, distinguishes it from censorship, and maps its scope. Anchors all subsequent technical and policy learning to a shared conceptual baseline.

Lesson 1 • Contextual and Nuanced Policy Clauses

Addresses satire, news reporting, education, and artistic expression exceptions. Teaches how to embed context-sensitive clauses without undermining enforcement clarity.

Lesson 2 • Writing Clear and Enforceable Rules

Covers plain-language drafting, specificity, and avoiding ambiguity in policy text. Students practise rewriting vague rules into actionable, testable standards.

Lesson 3 • Policy Versioning and Change Management

Explains how to update policies responsibly, communicate changes to users, and maintain version histories. Links policy evolution to emerging harm trends identified in Chapter 1.

Lesson 4 • Anatomy of a Community Guideline

Breaks down the structural components of effective platform policies. Teaches students to identify scope, definitions, prohibited behaviours, and enforcement consequences.

Lesson 1 • Consistency and Bias Mitigation

Identifies cognitive biases that distort moderation decisions and introduces calibration techniques. Connects consistent decision-making to the fairness principles established in Chapter 1.

Lesson 2 • The Moderation Decision Framework

Introduces a step-by-step decision process: identify, classify, contextualise, decide, and document. Provides the analytical backbone for all case evaluation work in this chapter.

Lesson 3 • Applying Decision Trees

Teaches construction and use of branching decision trees for common harm categories. Students build and test trees against sample cases to validate logic.

Lesson 4 • Handling Ambiguous and Edge Cases

Provides strategies for cases that fall outside clear policy coverage. Teaches escalation criteria, peer consultation, and provisional decision documentation.

Lesson 1 • Enforcement Action Taxonomy

Catalogues available actions from content removal and labelling to account suspension and legal referral. Establishes a shared vocabulary for enforcement used throughout the chapter.

Lesson 2 • Graduated Sanction Systems

Designs and implements strike-based and tiered sanction systems. Students map violation types to sanction tiers and define reset and rehabilitation pathways.

Lesson 3 • Proportionality in Enforcement

Applies the proportionality principle from Chapter 1 to action selection. Teaches matching sanction severity to harm level, intent, and recidivism history.

Lesson 4 • Emergency and Crisis Enforcement

Covers rapid-response enforcement during breaking news, coordinated attacks, and platform crises. Teaches temporary measures, war-room protocols, and post-crisis review.

Lesson 1 • Communication Transparency Reporting on Appeals

Covers how to aggregate and publish appeals data as part of platform communication transparency obligations. Connects individual case documentation from Chapter 3 to systemic accountability reporting.

Lesson 2 • Handling Repeat and Bad-Faith Appeals

Identifies patterns of abusive or bad-faith appeals and provides strategies to manage them. Protects reviewer capacity while maintaining legitimate access to the appeals process.

Lesson 3 • Writing Effective Enforcement Notices

Teaches clear, empathic, and policy-grounded user communication for enforcement actions. Reduces re-appeals and user confusion through precise, actionable notice language.

Lesson 4 • Principles of Due Process in Moderation

Translates legal due process concepts into platform practice: notice, opportunity to respond, and impartial review. Grounds appeals system design in the ethical principles from Chapter 1.

Lesson 1 • How Automated Moderation Works

Explains classifiers, hash-matching, natural language processing, and image recognition as moderation tools. Provides technical literacy needed to configure and evaluate automated systems.

Lesson 2 • Limitations and Failure Modes of AI

Catalogues common AI failure modes: context blindness, adversarial evasion, and language gaps. Prepares moderators to recognise when automated systems require human override.

Lesson 3 • Human-in-the-Loop Integration

Designs workflows where automated flags route to human reviewers at appropriate confidence thresholds. Balances speed and accuracy by defining clear handoff points between systems and people.

Lesson 4 • Configuring Automated Systems

Teaches threshold setting, allowlist and blocklist management, and confidence score interpretation. Students practise configuring a simulated classifier for a defined policy scope.

Lesson 1 • Designing QA Scorecards

Teaches construction of weighted scorecards that reflect policy priorities and harm severity. Students build a scorecard aligned to the enforcement taxonomy from Chapter 4.

Lesson 2 • Defining Moderation Quality

Establishes accuracy, consistency, speed, and policy adherence as the four pillars of moderation quality. Provides the measurement framework applied throughout the chapter.

Lesson 3 • Sampling and Audit Methodology

Covers random, stratified, and targeted sampling strategies for QA audits. Ensures audit samples are representative and statistically meaningful for performance conclusions.

Lesson 4 • Performance Dashboards and Reporting

Teaches construction and interpretation of real-time and periodic performance dashboards. Students identify trends, flag outliers, and generate actionable improvement recommendations.

Lesson 1 • Risk Assessment and Prioritisation

Applies risk frameworks to identify and prioritise the highest-impact harm vectors on a platform. Teaches resource allocation decisions grounded in harm probability and severity data.

Lesson 2 • Moderation Programme Governance

Defines governance structures including policy councils, escalation committees, and external advisory boards. Establishes accountability mechanisms that sustain programme integrity at scale.

Lesson 3 • Building a Moderation Programme Roadmap

Synthesises all course competencies into a structured programme roadmap with milestones, owners, and success metrics. Students present a complete roadmap as the capstone deliverable.

Lesson 4 • Measuring Programme-Level Impact

Extends QA metrics from Chapter 7 to programme-level outcomes: harm reduction, user trust, and policy effectiveness. Teaches longitudinal measurement and impact attribution methods.