HKU PHIL7002: Ethics: AI, Data and Algorithms — course overview with AskSia
PHIL7002 Ethics: AI, Data and Algorithms is one of the three 9-credit required core courses of HKU's Master of Arts in the field of AI, Ethics and Society, taught by the Department of Philosophy in the Faculty of Arts. The MA consists of six semester-long courses (three cores, three electives) and a 15-credit portfolio project, over one year full-time or two years part-time.
Sia generates PHIL7002 practice questions, walks through ethical implications of ai and theories step by step, and quizzes you on the material the component that weights most heavily.
Sharpen your argument
A firm's hiring algorithm rejects applicants from one district at twice the rate of others, and the firm says the model never uses district as an input. Which response applies the course best?
The syllabus's outcomes include the risks and abuses of algorithmic decision making in social contexts and the ethical concerns of large data sets.
Fairness and procedural justice are named in the syllabus as the tools for judging the outcome and the process.
Blanket rejection and unexamined acceptance both skip the normative analysis the course exists to teach.
One component decides 100% of your grade. Continuous; 100% coursework. This whole page is built around that.
Overview
What PHIL7002 is, and where it sits
PHIL7002 Ethics: AI, Data and Algorithms is one of the three 9-credit required core courses of HKU's Master of Arts in the field of AI, Ethics and Society, taught by the Department of Philosophy in the Faculty of Arts. The MA consists of six semester-long courses (three cores, three electives) and a 15-credit portfolio project, over one year full-time or two years part-time.
The syllabus sets out to give students an understanding of the broad ethical implications of artificial intelligence, the use of big data, and the role of algorithms in decision making, through theories and toolsets for thinking about the normative implications of predicting and mitigating the ethical risks they pose, including fairness and procedural justice. It gives special focus to the social, moral and economic effects of the widespread deployment of AI systems such as the large language models behind interfaces like ChatGPT.
By the end, students should be able to explain the ethical complexities of different forms of modern AI and deep learning, the ethical and privacy concerns of large data sets, the risks of adversarial attacks on otherwise harmless systems, and the risks and abuses of algorithmic decision making in social, political and interpersonal contexts. All instruction is in English and assessment is 100% coursework.
Always treat your own course outline and the exam timetable as authoritative.
Difficulty & time commitment
Is PHIL7002 hard, and how much time does it take?
PHIL7002 is manageable if you keep a weekly rhythm and treat the back half as the main event. The pattern is consistent: it starts gently and steepens, and the heaviest assessment is the part that separates grades.
The difficulty curve and the assessment weighting point the same way: the back half is harder and worth more. Front-loading effort there is the highest-return decision in the course.
Is this course for you
Who tends to do well, and who tends to struggle
You will likely do well if
- You can hold a normative theory and a technical mechanism in the same paragraph.
- You want to argue about specific systems, such as a language model or a hiring algorithm, not about AI in general.
- You read philosophy and are ready to read some machine learning.
You may struggle if
- You expect a technology-policy survey; this is applied ethics with philosophical standards of argument.
- You describe a harm without saying which theory makes it a wrong and to whom.
- You treat ChatGPT-style systems as the only case; the syllabus covers data, algorithms and adversarial attacks too.
- For each system you discuss, write its mechanism in three lines before writing its ethics in three pages.
- Keep a table of normative frameworks against the four outcome areas: AI forms, data sets, adversarial risk, algorithmic decisions.
- Draft one essay you would be willing to grow into the portfolio project.
Syllabus
The 10 topics, topic by topic
Topics from the public source. Their order is a content map, not an official weekly timetable or published exam weighting.
T1 · Ethical implications of AI, big data and algorithmic decision making
Syllabus descriptionThe three objects of the course and why each raises distinct ethical questions.
T2 · Theories and toolsets for normative analysis
Syllabus descriptionFrameworks for predicting and mitigating the ethical risks of AI, data and algorithms.
T3 · Fairness and procedural justice
Syllabus descriptionWhat fair treatment by an algorithm would require and how procedures can be just.
T4 · Effects of deploying AI systems
Syllabus special focusThe social, moral and economic effects of widespread AI, with large language models as the running case.
T5 · Large data sets: ethical and privacy concerns
Syllabus outcomeWhat collecting and using data at scale does to privacy and consent.
T6 · Adversarial attacks on otherwise harmless systems
Syllabus outcomeHow benign systems can be turned to harm and what that implies ethically.
T7 · Algorithmic decision making in social, political and interpersonal contexts
Syllabus outcomeRisks and abuses when algorithms decide about people.
T8 · Predicting and mitigating ethical risk
Syllabus descriptionUsing the normative toolsets to anticipate harms before deployment and to reduce them.
T9 · Forms of modern AI and deep learning techniques
Syllabus outcomeWhy different model types raise different ethical complexities.
T10 · Coursework: essays, problem sets and projects
AssessmentThe forms of assessment the MA lists programme-wide.
How it's assessed
Assessment structure
| Component | Weight | Format & timing |
|---|---|---|
| Coursework (programme-wide: may include discussion, participation, oral presentations, tests, research essays, problem sets, group work, written reports and design or outreach projects) | 100% | Coursework in the forms the programme lists (discussion, participation, oral presentations, tests, research essays, problem sets, group work, written reports, design or outreach projects); the split for this course is not published. Across the semester. Continuous; 100% coursework. |
- 100% coursework, which across the MA may include discussion, participation, oral presentations, tests, research essays, problem sets, group work, written reports and project work. Only passed courses earn credits; failing more than two courses in an academic year is a ground for discontinuation.
- No examination. Coursework is assessed against the course's stated outcomes: explaining the ethical complexities of modern AI, the concerns raised by large data sets, adversarial risks, and the abuses of algorithmic decision making.
This is a coursework course. Coursework carries 100% of the grade and the coursework (programme-wide: may include discussion, participation, oral presentations, tests, research essays, problem sets, group work, written reports and design or outreach projects) is the single heaviest piece at 100%, so steady work across the semester decides your result more than any one sitting. Continuous; 100% coursework.
Final exam timing: No examination. Confirm the exact date and venue on your exam timetable.
How to actually pass it
A weekly rhythm, two checklists, and the traps to avoid
The course rewards consistency over cramming, and practice over re-reading. Here is the loop that works, then what to have nailed before each exam.
The weekly loop
The mistakes that cost marks
Formula & concept sheet
The vocabulary and formulas you must own
- Algorithmic decision making
- Decisions about people produced or shaped by an automated procedure.
- Fairness
- The demand that an algorithm treat people or groups without unjustified disadvantage.
- Procedural justice
- The fairness of the process by which a decision is reached, independent of its outcome.
- Big data
- Data collected and processed at a scale that changes what can be inferred about people.
- Large language model
- An AI system trained on text at scale that generates language, such as those behind chat interfaces.
- Adversarial attack
- A deliberately crafted input that makes an otherwise harmless system behave harmfully.
- Privacy
- Control over information about oneself, strained by large data sets.
- Normative theory
- An account of what makes an action or practice right or wrong.
Where it fits
Prerequisites, related courses & why it matters
No course prerequisites; 9 credits; one of the three required core courses of the MA in the field of AI, Ethics and Society (one year full-time or two years part-time).
Your PHIL7002 study toolkit
Study the course with Sia, not just read about it
Use your permitted course notes with these study tools. Follow your course’s rules for AI and assessed work.
FAQ
Frequently asked questions
Is PHIL7002 hard?
Moderate on the six-factor rubric: no exam and no mathematics, but it is a 9-credit core marked on philosophical argument about technical systems.
What is the assessment?
100% coursework. The MA syllabus lists the forms assessment may take programme-wide; a course-level split is not published.
Do I need a technical background?
Admission to the MA is with a philosophy major or another major with relevant experience; the technical fundamentals are taught in the parallel core PHIL7001 Fundamentals of AI, Data and Algorithms.
Is it compulsory?
Yes. PHIL7001, PHIL7002 and PHIL7003 are the three required core courses (27 credits) of the MA in the field of AI, Ethics and Society.
Who teaches it?
The syllabus says the MA is taught by the Department of Philosophy; it does not name the course's teaching staff.
How does it connect to the portfolio project?
The 15-credit capstone portfolio builds on assignments and research done in the coursework, so essays written here can become the seed of the portfolio.
Study PHIL7002 with Sia
Practise explaining the course concepts and check your reasoning. Confirm assessed-work requirements in your current outline.
Start studying with Sia
