1. Home ›
  2. Course Calendar ›
  3. Aligning Values in AI Safety (Spring)

Aligning Values in AI Safety (Spring)

Thu, Apr 13

Guest Speaker: Ruyuan Wan

We began the semester discussing the different family of values. In today's session, we will discuss how those values could be aligned with AI technologies when discussing issues of safety and fairness. We have three main in-class learning goals. By the end of lecture today you will:

  1. Have an understanding of the concept AI value alignment.
  2. Consider the problem of AI bias and how it can lead to harmful outcomes like discrimination.
  3. Survey the different applications where AI technologies are being used, and pose questions about which ones are beneficial and which are harmful.

The slides for today's lecture.

Read This:

AI Value Alignment

  1. Bernard Marr, The Dangers Of Not Aligning Artificial Intelligence With Human Values, Forbes, 2022
  2. IBM Design for AI - Value Alignment
  3. Human-Centered Artificial Intelligence: Reliable, Safe & Trustworthy 
  4. Value-Sensitive Algorithm Design: Method, Case Study, and Lessons

AI Fairness

  1. Ellora Thadaney Israni, When an Algorithm Helps Send You to Prison, The New York Times, 2017

AI Accountability 

  1. IBM Design for AI - Accountability https://www.ibm.com/design/ai/ethics/accountability/
  2. AI accountability: Who's responsible when AI goes wrong?

AI Transparency 

  1. Why transparency matters in healthcare AI?

Do This:

Writing Reflection 10

Instructions for Writing Reflection 10.

This writing reflection is due on 4/14 at 12pm.

This Week's Dialogue Group Meeting

Find at least one hour to meet with your group to discuss the prompt of the week: What concerns about AI are legitimate at the present moment?