Lecture 12: Ethics Discussion

HTML Slides | PDF Slides

Generative AI, ethics, and policies

COMP 4630 | Winter 2026 Charlotte Curtis

Guiding questions for discussion

• What are some ways that AI can be helpful? Harmful?
• What are some ways that AI is used?
  • Axes: intent (malicious <-> innocent), impact (harmful <-> helpful)
• What concerns do you have about AI in general?
  • What about in education, specifically at MRU?
• What would you like to see in an AI policy at MRU?

Case study: vibe coding

Costs* of:

• Voice to text
• Code generation
• Debugging
• at ~3 Wh/request

April fools?

• Yesterday (March 31st, 2026), Claude Code’s source code was accidentally leaked
• It’s an interesting code base to browse through, notably:
  • April Fool’s day buddy sprite
  • Negative keyword matching
  • Undercover mode

Impact

• How do you, as CS majors, feel about these trends?
• Have you tried using agentic coding tools?
• What is the true cost of agentic coding?
  • A recent paper attempts to quantify token usage by task
  • Environmental impact, energy consumption, financial cost

Coming up next

• Easter break, no class on Monday
• Wednesday: One last journal club presentation (YOLO) and a smattering of additional topics (autoencoders, object detection, image generation?)
• Monday: last day of class! Assignment 3 results, project checkpoint discussions. Maybe some kind of treats.