CS 6603: AI, Ethics, and Society
Review and Retrospective
Disclaimer
The views and opinions expressed in this post are solely my own and do not reflect those of Georgia Tech, the OMSCS program, or any affiliated instructors, TAs, or staff.
Instructor: Dr. Ayanna Howard
Semester: Fall 2024
Overall Rating: 4.1 / 10 👎
✅ Pros
- Introduces fairness and bias‑mitigation
- Touches on some U.S. and international regulations around algorithmic fairness, bias, and data privacy
- You could feasibly complete the coursework in far less time than the schedule suggests
❌ Cons
- Falls short of its responsibility to keep pace with the evolving AI/ML ethical issues impacting society today
- Examples and datasets are dated (some pre‑2016) uundermining course relevance
- Most assignments seem like rote busy‑work
- The lectures are dumbed down to a point that they read closer to a high‑school survey than a graduate‑level course
🕒 Time Commitment
Plan on 1 hour per week for lectures and assignments.
📝 Grade Breakdown
| Component | Weight | Notes |
|---|---|---|
| Homework Projects | 40 % | Four applied mini‑projects |
| Written Critiques | 10 % | Two short papers |
| Mid‑Term Exam | 10 % | Closed‑book, proctored via Honorlock |
| Final Project | 15 % | Bias‑mitigation case study |
| Final Exam | 10 % | Closed‑book, proctored via Honorlock |
| Class Discussions / Exercises | 15 % | Ed threads, in‑video quizzes |
✍️ Assignments
FB Homework
Audit data‑collection & privacy in social media.
Stats 101 Notebook
Replicate Anscombe’s Quartet & sampling‑bias demos.
AI/ML Parts I & II
Explore bias in word embeddings, facial recognition, and predictive policing.
Fairness Toolkit Lab
Hands‑on with IBM AI Fairness 360 and What‑If Tool.
📖 Exams
Two very straighforward proctored exams.
🗂 Final Project
You’re asked to quantify bias in a dataset and demonstrate at least one mitigation technique. Then argue its societal trade‑offs in a 10‑page report (15 % of grade).
📚 Course Content
| Module | Topics | Details |
|---|---|---|
| 1 | Data, Individuals & Society | Case‑studies on loan denial, email leaks |
| 2 | “BS of Big Data” & Statistics Basics | Anscombe’s Quartet, smoking‑bias design |
| 3 | Fairness in AI/ML | Bias in word embeddings & facial recognition |
| 4 | Bias Mitigation & Futures | AI Fairness 360, What‑If Tool, final wrap‑up |
💬 Participation & Interaction
Course content is very straightforward with no need for office hours or Ed Discussion. You’re sometimes asked to weigh in on other student’s posts.
💭 Final Thoughts
CS 6603 tackles an undeniably important subject, yet the execution lags behind the AI landscape. Outdated examples and an almost child‑like delivery style left me wanting more depth and currency. A refresh of materials and tone could easily turn this from a 4.1/10 into a must‑take course.