AI and Data Ethics

At the end of the course, you will earn a Certificate of Completion from Spirit in Projects.

Are you ready for innovation while taking on social responsibility? Then you might want to engage with ethical dilemmas and measurement possibilities for the quality of Artificial Intelligence. If you want to learn more about obligations in connection with AI applications and actively participate in current discussions, then look forward to this training. Together we will analyze case studies, discuss the global impacts of AI development, and intensively examine the requirements of the EU AI Act and its consequences. Additionally, you will learn about the specifics of project management for AI projects and simulate a moot court.

Advising

Objectives

  • Understand AI ethics and apply it in practice
  • Develop ethical guidelines for AI use in the enterprise
  • Manage data protection requirements and bias challenges
  • Discuss and implement relevant legal frameworks
  • Analyze best practices for ethical AI development
  • Successfully plan and manage AI projects
  • Implement MLOps and model governance

Target groups:

Business Analyst, Requirements Engineer, AI Expert, Project Manager, Project Lead, Demand Manager, Portfolio Manager, IT Project Manager, Product Owner, Compliance Manager, Legal Counsel, Data Protection Officer and anyone who wants to engage with AI ethics and AI project management.

Termine anzeigen

Syllabus

1. Fundamentals

  • Ethics and AI
  • Principles of ethical AI development
  • AI ethics dilemmas in practice
  • Ethical reasoning
  • Responsible AI

2. Data Protection

  • Overview of GDPR
  • Principles of data processing
  • Rights and obligations
  • Role of the data protection officer
  • Challenges in AI systems
  • Privacy by Design
  • Anonymization and pseudonymization

3. EU AI Act

  • Overview of the EU AI Act
  • Classification of AI systems: Risk categories
  • Prohibited AI practices
  • High-risk AI systems
  • Responsibilities and obligations
  • Requirements for transparency and documentation
  • Consequences of non-compliance
  • Practical implementation in enterprises

4. Bias and Fairness

  • What is bias in AI systems?
  • Types of bias: Data bias, algorithmic bias, human bias
  • Impacts of unfair AI systems
  • Fairness metrics and definitions
  • Methods for bias detection
  • Mitigation strategies
  • Fairness-aware Machine Learning
  • Case studies on bias incidents

5. Quality in AI Systems

  • Quality in AI: What does it mean?
  • Dimensions of quality
  • Metrics for various AI systems
  • Independent testing procedures
  • Validation and verification
  • Challenges of quality assurance
  • Recommendations and trends
  • Continuous evaluation

6. An International Perspective

  • EU: AI Act and digital strategy
  • USA: AI Executive Order and regulatory approaches
  • China: Social scoring and AI regulation
  • Qatar: AI strategy and development
  • United Arab Emirates: AI innovation and governance
  • Global standards and harmonization
  • Comparison of regulatory approaches

7. Copyright and Legal Challenges

  • Overview of copyright law
  • Exceptions for AI training
  • Copyright and digital media
  • AI-generated content: Who owns it?
  • Current court cases and precedents
  • Liability issues in AI systems
  • Contract design for AI projects

8. Project Management of AI Projects (NEW)

  • AI project lifecycle: From idea to deployment
  • Specifics of AI projects vs. traditional IT projects
  • Iterative development and experimentation
  • Proof of Concept (PoC) design and evaluation
  • Data Science workflow: CRISP-DM for AI projects

9. Roles and Team Organization (NEW)

  • Roles in AI teams: Data Scientists, ML Engineers, AI Product Manager
  • Responsibilities and interfaces
  • Interdisciplinary collaboration
  • Skill requirements and training
  • External vs. internal expertise
  • Stakeholder management in AI projects
  • Budget, resources, and KPIs

10. MLOps and Model Governance (NEW)

  • What is MLOps? DevOps for Machine Learning
  • CI/CD pipelines for ML models
  • Model versioning and experiment tracking
  • Model registry and deployment
  • Monitoring and retraining
  • A/B testing and model evaluation in production
  • Tools: MLflow, Weights & Biases, Kubeflow

11. Moot Court

  • Simulation of a court case on AI ethics of a project
  • Role assignment and preparation
  • Execution and argumentation
  • Reflection and lessons learned
Spirit in Projects