Graduate course 11118BLG001, asynchronous, 14 weeks

Machine Ethics and Artificial Intelligence Safety

Prof. Dr. Utku Kose, Süleyman Demirel University

A 14-week asynchronous graduate course with lecture notes, interactive labs, Colab notebooks and weekly self-assessment

Each week combines lecture notes, an interactive lab that runs in the browser, a Colab notebook with self-checking exercises and a self-assessment with confidence ratings. Progress is stored only in this browser.

This course is updated in line with current developments in the field. Last update: September 2026.

These materials were prepared for the graduate course 11118BLG001 Machine Ethics and Artificial Intelligence Safety at Süleyman Demirel University. They are openly available: Anyone may use and adapt them in courses of related scope, with attribution, under the CC BY 4.0 licence for content and the MIT licence for code.

Weekly modules

  1. 01Foundations of Machine Ethics and AI SafetyLectureInteractive labColab notebookPDF notesNot started
  2. 02Normative Ethics as Decision ProceduresLectureInteractive labColab notebookPDF notesNot started
  3. 03Implementing Artificial Moral AgentsLectureInteractive labColab notebookPDF notesNot started
  4. 04Moral Dilemmas, Pluralism and Social ChoiceLectureInteractive labColab notebookPDF notesNot started
  5. 05Algorithmic Fairness I: Harms, Metrics and ImpossibilityLectureInteractive labColab notebookPDF notesNot started
  6. 06Algorithmic Fairness II: Mitigation and Representational HarmsLectureInteractive labColab notebookPDF notesNot started
  7. 07Specification Problems: Reward Hacking, Side Effects and Goodhart's LawLectureInteractive labColab notebookPDF notesNot started
  8. 08Robustness: Adversarial Examples, Poisoning and BackdoorsLectureInteractive labColab notebookPDF notesNot started
  9. 09Uncertainty, Calibration and Distribution ShiftLectureInteractive labColab notebookPDF notesNot started
  10. 10Interpretability and Transparency for SafetyLectureInteractive labColab notebookPDF notesNot started
  11. 11Aligning Language Models: Preferences, Feedback and OversightLectureInteractive labColab notebookPDF notesNot started
  12. 12Security of LLM Applications: Prompt Injection, Jailbreaks and Data ExtractionLectureInteractive labColab notebookPDF notesNot started
  13. 13Advanced Risks: Goal Misgeneralization, Power-Seeking and DeceptionLectureInteractive labColab notebookPDF notesNot started
  14. 14Governance, Standards and AuditingLectureInteractive labColab notebookPDF notesNot started

Midterm and final capstones

  1. MTMidterm Capstone: Ethics, Fairness and Specification in PracticeCapstone pageEnd of Week 7
  2. FNFinal Capstone: Building and Auditing a Safer AI SystemCapstone pageEnd of Week 14