Understanding AI

All the analytics integrations you will ever need.

3 min

AI Literacy : Policy and Practice

This lesson consolidates and summarizes the core technical, conceptual, and ethical understanding of AI necessary for anyone involved in its governance. It clarifies that AI literacy extends beyond basic definitions, encompassing an appreciation for AI's diverse capabilities, inherent limitations, underlying mechanisms (e.g., data's role, model lifecycle), and the broader ethical implications, all crucial for informed decision-making in governance roles.

EU AI Act, OECD AI Principles, NIST AI Ris

10 min

Governing AI: From Frameworks to Function

Outlines the phases of AI governance from design to post-deployment monitoring. Offers examples of effective ongoing governance.

NIST AI Risk Management Framework, ISO/IEC 27001

10 min

Principles for Responsible AI

Delves into core ethical principles guiding responsible AI, such as fairness, accountability, transparency, privacy, safety, and human oversight. It frames societal impact within these principles, discussing how adherence minimizes harm and maximizes benefit to individuals and communities.

OECD AI Principles, UNESCO Recommendation on the Ethics of AI, IEEE 7000 Series, AI Bill of Rights.

7 min

Applied AI: From Lab to Industry

Applies AI concepts to healthcare, finance, and marketing. Discusses risk tradeoffs and governance concerns across sectors.

OECD AI Principles, EU AI Act

7 min

Explainable AI: Unpacking Transparency

Introduces the concept of Explainable AI (XAI), why it's crucial for understanding complex AI models, and various techniques to provide transparent and interpretable insights into their decision-making processes.

EU AI Act (Transparency), NIST AI Risk Management Framework (Explainability), ISO/IEC 23894.

7 min

Hallucination and Bias

Demystifies hallucinations in AI systems and discusses risks like misinformation, bias, and loss of trust.

ISO/IEC 27001, GDPR

5 min

How Models Learn (and Fail)

Explains common issues where models either perform too well on training data but poorly on new data, or are too simplistic to capture underlying patterns, affecting real-world accuracy.

NIST AI Risk Management Framework

5 min

Synthetic Data: Purpose and Applications

Defines synthetic data and how it helps train AI models ethically and efficiently, especially where real data is sensitive.

GDPR, EU AI Act

7 min

AI Bias: Data to Model

Details the various ways bias can enter AI systems through data, from initial collection to its use in training models, and outlines methods for detection and reduction.

OECD AI Principles, ISO 42001

7 min

Types of AI

Covers narrow, general, generative, and superintelligent AI types. Highlights use cases across industry.

OECD AI Principles, NIST AI Risk Management Framework

10 min

LLMs: Limits and Governance

Explains the architecture behind LLMs and their policy implications. Highlights governance risks in content generation and automation.

OECD AI Principles, EU AI Act

7 min

Narrow vs. General AI

Explores the fundamental differences between AI designed for specific tasks and AI aiming for human-like intelligence across domains.

OECD AI Principles, IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems

10 min

AGI: Hype vs. Reality

Explores the elusive nature of Artificial General Intelligence (AGI), critically examining its definitions, the current state of research, and the inherent weaknesses in establishing clear benchmarks for human-level AI. It addresses the gap between public perception and scientific reality.

None (Focus is on conceptual understanding and critique rather than formal frameworks)

7 min

AI Lifecycle: Data to Decisions

Explains data collection, model training, deployment, and monitoring. Ties each stage to compliance and security standards.

NIST AI Risk Management Framework, ISO/IEC 27001 (data security)

7 min

ML and Neural Networks

Covers neural networks and deep learning. Real-world uses in NLP, facial recognition, and image analysis.

NIST AI Risk Management Framework

7 min

What Is an AI System?

Breaks down AI into models, algorithms, and data. Examples from finance and healthcare show AI systems in action.

EU AI Act (system classification)

7 min

Types of Machine Learning

Clarifies how AI systems learn and operate through distinct methods: learning from labeled data, finding patterns in unlabeled data, or trial-and-error with feedback.

None (These are fundamental ML paradigms, not specific frameworks)

5 min

What Is Machine Learning?

Clarifies how machine learning differs from traditional programming. Includes use cases like fraud detection and content recommendations.

NIST AI Risk Management Framework

5 min

What Is AI?

Defines AI in relation to machine learning and deep learning. Explains key distinctions using real-world tools like smart assistants.

OECD AI Principles

Understanding AI

Concepts, Capabilities, and Core Components

1.01

What Is AI?

A short introducion to the core ideas behind artificial intelligence.

Key Concepts

Definition of AI and its relationship with machine learning and deep learning.

Learning Time

5 min

1.02

What Is Machine Learning?

Understand how machines learn patterns instead of following rules.

Key Concepts

The difference between machine learning and traditional programming.

Learning Time

5 min

1.03

Types of Machine Learning

ML Types: Supervised, Unsupervised, Reinforcement

Key Concepts

Labeled data; unlabeled data; reward signals; iterative learning.

Learning Time

7 min

1.04

ML and Neural Networks

Explore the building blocks of modern AI systems.

Key Concepts

Basics of neural networks, deep learning, and their role in AI systems.

Learning Time

7 min

1.05

What Is an AI System?

Understand the basic components that make up an AI system.

Key Concepts

Components of an AI system: data, models, and algorithms.

Learning Time

7 min

1.06

AI Lifecycle: Data to Decisions

Follow the AI lifecycle from raw data to real-world outputs.

Key Concepts

The lifecycle of AI: data collection, model training, deployment, and monitoring.

Learning Time

7 min

1.07

Types of AI

Learn the spectrum from narrow to superintelligent AI.

Key Concepts

Overview of Narrow AI, Generative AI, General AI, and Superintelligent AI—their capabilities and distinctions.

Learning Time

7 min

1.08

Narrow vs. General AI

The spectrum from focused tasks to human-like intelligence.

Key Concepts

ANI vs. AGI; task-specific vs. generalized intelligence; current capabilities.

Learning Time

7 min

1.09

LLMs: Limits and Governance

What large language models can and can’t do — and why it matters.

Key Concepts

Understanding large language models (LLMs) and their architectural limits.

Learning Time

10 min

1.1

AGI: Hype vs. Reality

Questioning the existence and definition of Artificial General Intelligence.

Key Concepts

Artificial General Intelligence (AGI) vs. Narrow AI (ANI); definitional challenges (e.g., common sense, consciousness, generalization); current AI capabilities (e.g., LLMs) and their limitations; the Turing Test and its critiques; the "AI Winter" phenomenon.

Learning Time

10 min

1.1

Hallucination and Bias

When AI makes things up — and why it’s a governance issue.

Key Concepts

Understanding hallucinations and fabrications in AI-generated content.

Learning Time

7 min

1.11

AI Bias: Data to Model

Identifying and mitigating data-driven bias in AI systems.

Key Concepts

Data collection bias; historical bias; selection bias; measurement bias; label bias; data auditing; data augmentation; re-weighting.

Learning Time

7 min

1.13

Synthetic Data: Purpose and Applications

Description of synthetic data and how it differs from anonymized data.

Key Concepts

Definition of synthetic data and how it differs from anonymized data.

Learning Time

5 min

1.14

How Models Learn (and Fail)

Optimizing AI models for accuracy and generalization.

Key Concepts

High training accuracy/poor test performance; high errors on all data; model complexity; generalization.

Learning Time

5 min

1.14

Explainable AI: Unpacking Transparency

Demystifying AI decisions for trust and accountability.

Key Concepts

Interpretability vs. explainability; local vs. global explanations; XAI techniques (e.g., SHAP, LIME, saliency maps, feature importance); post-hoc explanations.

Learning Time

7 min

1.15

Applied AI: From Lab to Industry

Applications of AI across healthcare, finance, and marketing.

Key Concepts

The ethical considerations and risks of using AI in these industries.

Learning Time

7 min

1.16

Governing AI: From Frameworks to Function

End-to-end oversight for responsible AI.

Key Concepts

Steps in designing, implementing, and maintaining AI governance frameworks.

Learning Time

10 min

1.17

AI Literacy : Policy and Practice

Defining the essential AI knowledge required for informed oversight and effective governance.

Key Concepts

Technical, conceptual, and ethical concepts within the topic of AI necessary for anyone involved in its governance.

Learning Time

3 min

1.18

Principles for Responsible AI

Guiding AI development for human well-being and responsible innovation.

Key Concepts

Beneficence, non-maleficence, autonomy, justice, accountability, transparency, privacy, security, human oversight. Societal impacts (e.g., job displacement, misinformation) as ethical challenges.

Learning Time

10 min

The EU AI Act

The EU's Landmark AI Regulation

2.01

Framing the EU AI Act

Regulatory context, policy objectives and practical scope

Key Concepts

Regulatory purpose, scope of application, global context

Learning Time

2.02

Essential concepts

What and who is covered by the EU AI Act

Key Concepts

AI system, high-risk AI systems, general-purpose AI models, providers, deployers

Learning Time

2.03

Scope of Application

When the EU AI Act applies and what is outside its scope

Key Concepts

Placing on the market, putting into service, affected persons

Learning Time

2.04

The risk-based approach of the EU AI Act

Identifying the levels of risk of AI and their significance

Key Concepts

Prohibited AI practices, high-risk AI systems, transparency risks

Learning Time

2.05

Implementation timelines and strategic planning

Compliance deadlines and internal readiness planning

Key Concepts

Key dates, AI literacy

Learning Time

2.06

General-purpose AI models

GPAI concept and applicable requirements

Key Concepts

GPAI model, GPAI model with systemic risk, Code of practice

Learning Time

2.07

Obligations of providers of high-risk AI systems

Applicable requirements and practical steps to comply

Key Concepts

Risk management system, data governance, human oversight, quality management system

Learning Time

2.08

Obligations of deployers of high-risk AI systems

Applicable requirements and practical steps to comply

Key Concepts

Human oversight, data protection impact assessment, fundamental rights impact assessment

Learning Time

2.09

Conformity assessments

Effect of conformity assessments, applicable requirements and practical steps to comply

Key Concepts

Conformity assessments, notifying authorities, conformity assessment body, standards, certificates, registration

Learning Time

2.1

Regulatory governance, monitoring and enforcement

Governance bodies, market surveillance and enforcement powers and penalties

Key Concepts

AI Office, European Artificial Intelligence Board, national competent authorities, market surveillance

Learning Time

2.11

Global alignment and comparisons

Position of the AI Act in relation to international frameworks

Key Concepts

Global responsible AI principles, AI regulatory sandboxes

Learning Time

2.12

Privacy governance and GDPR transferable practices

Deployment of GDPR compliance strategies and practices in the context of the AI Act

Key Concepts

Controller/processor v. deployer/provider, data management, impact assessments, regulatory engagement

Learning Time

AI Governance In Practice

Implementing Principles, Policies, and Oversight