AIGP Certification Prep Course: IAPP Official Guide 2026

Define artificial intelligence as machines performing tasks that normally require human intelligence, with technology, autonomy, human involvement, and probabilistic outputs shaping governance.

Identify the five core traits of AI systems—adaptivity, autonomy, complexity, probabilistic behavior, and context awareness—and explore their governance, risk, and oversight implications.

Explore how AI automates, augments, and amplifies decision making, and learn governance guardrails like transparency, traceability, and human override to prevent flawed outcomes.

Apply the OECD framework to classify AI by function, context, data and inputs, human involvement, and technical autonomy to guide governance, audits, and responsible use.

Explore ANI, AGI, and ASI with real-world examples like Siri, Google Translate, and ChatGPT. Examine their capabilities, limitations, and ethical considerations as AI moves from ANI to AGI and beyond.

Explore the three core machine learning paradigms—supervised, unsupervised, and reinforcement learning—and see how labeled data, unlabeled data, and rewards guide model predictions and real-world applications.

Explore how deep learning builds abstract representations through neural networks, backpropagation, and architectures like CNNs and transformers, while addressing governance, explainability, and ethical deployment.

Explore how AI operates within sociotechnical systems, integrating technical, organizational, and human layers to shape governance, risk, and trust in real-world deployments.

Explore the human, organizational, and technical layers that govern real-world ai systems. Learn how governance, data choices, and workflow decisions shape trust, bias, and system safety across sociotechnical contexts.

Explore accountability in complex AI systems, from causal and moral diffusion to legal and remedial responsibility, and learn accountability by design and algorithmic impact assessment to close the accountability gap.

Apply the OECD AI classification framework to govern AI systems by assessing function, context, and impact, enabling risk-based, proportionate governance across sectors.

Apply the OECD classification framework to real-world AI in healthcare, finance, and policing, highlighting governance needs, bias testing, and ongoing monitoring.

Explore how open source AI models and cloud computing ecosystems democratize access to powerful AI tools while highlighting governance challenges and adaptive oversight.

Examine how competitive pressure drives ai acceleration, highlighting first mover and fast follower dynamics, talent and investor competition, regulatory races, and governance strategies like governance by design and competitive governance.

Navigate the geopolitics of AI innovation that shape governance decisions, highlighting US–China rivalry, EU regulation, and global south actors, and explore cooperative governance amid fragmentation and competition.

Map the landscape of AI-induced harms by examining impacts on individuals, groups, society, organizations, and ecosystems. Learn to classify risks using harm taxonomies for responsible AI governance.

Leverage harm taxonomies to organize AI risks into categories, drawing on the panoptic model and Calo's subjective and objective harms, plus Citron and Salovey's physical, reputational, and economic harms.

Examine how data-driven AI biases—including implicit, sampling, temporal, and overfitting—cause discrimination in hiring and housing, and learn to curb them through diverse teams and rigorous data curation.

Delve into AI privacy concerns, from training data and de-identification limits to data appropriation and inference. Emphasize transparency and governance to protect individual rights in the age of AI.

Explore how AI-related societal harms threaten democracy and public safety, including disinformation, deep fakes, and lethal autonomous weapons, and identify safeguards for governance professionals.

Address organizational and environmental harms through AI governance, covering reputational, cultural, economic, and legal risks. Training large models emits carbon dioxide and consumes energy and water, demanding greener governance.

Assess the operational and business risks of running ai systems, including costs, data security, vendor lock-in, and intellectual property concerns.

Identify and mitigate privacy risks across the AI lifecycle, including data persistence, repurposing, spillover, and informed consent, by integrating privacy by design and safeguarding data rights.

The five OECD principles for trustworthy AI establish inclusive growth, human-centered values and fairness, transparency and explainability, robustness and safety, and accountability to guide responsible AI development and policy.

Translate OECD principles into a practical seven-issue AI ethics checklist: lawfulness, safety, bias, transparency, data consent, human in the loop, accountability.

Formally adopt and champion ethical principles across the organization. Create a cross-functional, demographically diverse AI ethics committee to oversee high-risk projects and tailor policies with measurable impact.

Build a culture of ethical AI by aligning legal, compliance, and cross-functional teams around transparent, equitable design, explainability, privacy, data governance, minimization, consent, and responsible data use.

Explore the four characteristics of trustworthy AI: human centric, accountable, transparent and explainable, and privacy enhanced, and learn how they create reliable, fair, and safe AI systems.

Embed trustworthy AI into your operating model daily. Operationalize a five-part framework with risk management, checks, safety testing, and human oversight to ensure privacy, security, and accountability.

Foster a culture of responsible AI as a governance discipline, with a shared language, inclusive practices, and ongoing training. Highlight customer value and trust to motivate safe, fair AI.

Tailor ai governance to your organization by assessing six measurements—size, maturity, industry, products, objectives, and risk tolerance—and design a custom governance program that fits.

Identify the three roles—developers, deployers, and users—and explain their distinct responsibilities, documentation needs, and feedback loops that drive governance across the ai lifecycle.

Translate strategy into formal policies that enable oversight and accountability across the AI life cycle, covering use case assessment, risk management, ethics by design, data use, and deployment.

Identify key stakeholders in ai governance across policy, legal, privacy, and ethics, plus builders and users of ai systems. Align architecture, risk, and operations teams for practical governance.

Explore how cross-functional collaboration among researchers, data scientists, AI ML engineers, and non AI engineers strengthens AI governance by turning ethical principles into practical, reliable product decisions.

Identify AI champions and secure early leadership buy-in for governance; frame responsible AI as a competitive differentiator, outline costs and legal needs, and present proactive, solutions-oriented governance.

Leverage existing security and privacy processes to build a formal AI governance structure, start slowly, define responsibilities, document decisions, and appoint an executive champion while exploring centralized or decentralized models.

Explore centralized, decentralized, and hybrid AI governance models and how they align with your organization's culture. Define roles and responsibilities to ensure consistent, responsible AI development.

Identify and engage a diverse group of stakeholders early in AI governance, including leadership, privacy, and end users, to build a shared business case, a risk framework, and training plans.

Develop AI literacy across your organization by tailoring training and awareness to your AI use, governance, and policies, covering AI terminology, AI strategy, AI governance, privacy, fairness, and human agency.

Align your ai risk strategy with existing security, privacy, and business risk plans to avoid gaps. Conduct a risk analysis, implement mitigations, and document responsibilities in a preliminary analysis report.

Establish practical AI assessment processes using external frameworks like the NIST AI RMF and ISO standards, and adapt them to your context to prioritize fairness, bias, transparency, and safety.

Assess AI risk by analyzing context: ownership and operator, industry and use case, social impact, timing, and jurisdiction, then apply context-specific standards for responsible governance.

Evaluate whether your AI achieves its business purpose and planned uses, manages harms and data quality, tests functionality and metrics, and gauges third party risks to ensure safe outcomes.

Master the NIST AI Risk Management Framework (RMF) and ARIA program to build a structured, repeatable process for trustworthy AI risk, guided by governance, map, measure, and manage.

Outline the AI system development life cycle with four phases—plan, design, develop, implement—emphasizing governance, data handling, testing, deployment, monitoring, and ethical considerations for responsible AI projects.

Define the AI project by identifying the business problem and use case in the plan phase, then classify, predict, or recommend to prioritize impactful AI opportunities.

Apply three lines of defense governance to AI development to ensure human oversight, with clear roles for real-time operational responsibility, routine audits, feedback and appeal mechanisms, and a kill switch.

Apply risk assessment strategies throughout AI development using the probability-severity harms matrix, risk mitigation hierarchy, stakeholder mapping, use case evaluation, benchmarking, and pre-deployment pilots.

Design the system architecture and establish performance measures to choose an AI model by balancing accuracy and interpretability. Define success with thresholds, baselines, and continuous monitoring for fairness.

Train, test, and validate with representative data to ensure the model generalizes and meets goals. Document the model version and data used for transparent governance with regulators.

Evaluate whether you have the right data to make your AI usable, assess data accessibility and format, and embed privacy laws, data localization, and regulatory disclosures from the start.

Assess data quality by ensuring accuracy, representativeness, and bias to avoid garbage in, garbage out, and classify data as structured, unstructured, semi-structured, or static versus streaming for effective data wrangling.

Apply data preparation techniques from cleansing and labeling to anonymization and data minimization, guided by privacy enhancing technologies like differential privacy and federated learning.

Collaborate with domain experts to select the most important features, perform feature engineering, and maintain consistent, lean features for training and testing to improve performance and explainability.

Assess third-party AI tools via adapted procurement and vendor screening, establish AI-specific policies and inventory management, and monitor both integrated and employee-use tools with the NIST AI risk management framework.

Evaluate vendor and open source agreements by scrutinizing data rights, privacy, security, bias, technical specs, and use terms to manage risk and ensure production readiness.

Conduct a readiness assessment before production to verify model card transparency and conformity. Implement continuous monitoring, audits, threat modeling, security testing, maintenance, retraining, and public disclosures for a safe deployment.

Assess the deployed AI model through periodic performance, reliability, and safety checks, using audits, red teaming, threat modeling, and security testing with human oversight throughout its lifecycle.

Manage model drift and iteration to keep models accurate as data changes, and use retraining with a baseline to maintain governance, compliance, and alignment with business goals.

Treat AI failures as incidents and activate your incident response plan. Report, document, and communicate mitigation to relevant stakeholders while assessing third-party risks and preserving a human shutdown option.

Disclose AI use across deployments, audit incidents, and implement governance tools to ensure transparency and accountability. Learn notices, incident response, and automated compliance to build trust.

The EU AI act establishes a harmonized, risk-based regulation to govern AI development and deployment in the EU, balancing safety, transparency, and innovation while safeguarding fundamental rights.

The eu ai act has extraterritorial reach, applying to any organization using ai affecting eu citizens. It sets a global standard guiding compliance and supply chains.

Learn exemptions to the EU AI Act, including national security and defence, low risk research, exemptions for SMEs and startups, and open source AI considerations.

Define four main roles in the eu ai act: provider, deployer, importer, and distributor. Explore their respective responsibilities across the ai life cycle.

Explore the EU AI Act’s four risk levels in a risk-based framework, from prohibited to minimal risk, with key obligations to process, document, and audit AI use.

Explore the EU AI act’s prohibited AI systems, including social credit scoring, emotion recognition, manipulation, untargeted facial image scraping, biometric categorization, and real-time identification bans in public spaces.

Explore high-risk ai systems under the EU AI act, focusing on conformity assessments and safety before market, with two subcategories: product safety components and a list of high-risk areas.

Identify and implement EU AI act's high-risk obligations by establishing a risk management system, data governance, documentation, and human oversight throughout the AI life cycle to ensure safety and transparency.

Deployers of high-risk ai must complete a fundamental rights impact assessment, ensure proper use, enable human oversight, monitor performance, report issues, and maintain logs for risk management.

Explore limited risk ai systems under article 50, with transparency and user notification, and minimal risk ai with voluntary codes of conduct, plus deepfakes disclosure and consent for emotion recognition.

Discover how general purpose AI models (GPI) are regulated, detailing baseline obligations like transparency and copyright, plus higher risk rules, adversarial training, and data governance.

Explore the governance and enforcement structure of the EU AI act, including the central AI office, the AI board, an advisory forum, and sectoral regulators enforcing it.

Explore the eu ai act penalties and its phased enforcement timeline, including prohibitions for unacceptable risk ai and fines up to €35 million or 7% of global turnover.

Navigate the fragmented U.S. AI regulatory landscape from the White House AI Bill of Rights blueprint to executive orders and state and city laws, with FTC, CFPB, and OMB guidance.

Develop a flexible AI governance framework tailored to your organization. Define principles with stakeholders, set risk tolerance, consider jurisdictions and industry, and align use cases with your strategy and resources.

Discover the HUD Aria framework, a human rights based AI governance approach grounded in Council of Europe standards, built around eight principles for AI impact assessments.

Explore how the GDPR governs AI through article 22 on automated decision making, consent, and data subject rights, ensuring transparency, human oversight, and responsible AI governance.

Navigate GDPR and LGPD rules for sensitive data in AI, balance data minimization with bias testing, and apply strategies like direct collection, proxies, or consent-driven data use.

Data controllers must uphold GDPR processing principles, conduct DPIAs, protect data subject rights, and manage cross-border transfers when using AI.

Apply existing sector laws to AI and AI-based products, recognizing that AI operates within the same regulatory contexts and requires controls for both updated functions and new processes.

Explore how copyright, patent, fair use, and data licensing intersect with AI, including ownership of AI outputs, inventor status, and governance of model components and liabilities.

Explore how non-discrimination laws apply to AI, addressing bias in training data, privacy constraints, and regulatory guidelines across healthcare, insurance, employment, credit, and housing to promote fairness.

Explore how consumer protection and product liability laws apply to ai, with the ftc enforcing unfair or deceptive practices. Examine challenges in attributing harm to autonomous ai systems.