Adventures in Ethical Computing
summary
Explains how bias in software and hardware can cause real-world harm, citing examples like ML gender inference in translation, airbags harming short drivers/children, and VW emissions-cheating. Reviews academic literature and case studies to classify types of bias and recommends practical ways engineers can reduce bias in systems to improve user experiences and limit ethical/legal risk.
AI-written summary Β· about
chapters (15)
- 0:00Start
- 3:24WHAT'S A COMPUTER
- 4:12Ethics
- 5:09IS TRUTH GOOD
- 5:30YOU SHOULD NOT LIE
- 6:54CHRISTOPHER WYLIE
- 8:54AIR BAGS
- 12:38NEARLY 90% VEHICLES MADE BEFORE 1998
- 13:08WE CAN ALWAYS DO BETTER
- 15:00SYSTEMS INHERIT THE BIAS OF THEIR CREATORS
- 15:23THE SHIRLEY CARD
- 19:18THIS IS THE RESULT OF UNCONSCIOUS BIAS
- 20:31THE IRON RING
- 22:10CHANGE STARTS WITH 1 PERSON
- 23:08REFLECTIVE EQUILIBRIUM
related talks
47mThe Ethics of Software - some practical considerations
50mEthical Machine Learning
16mExistential Engineering or How Ethics Precedes Error Handling
37mEthics in Software Engineering - Should We Care
51mDigital Discrimination: Cognitive Bias in Machine Learning
37mDigital Discrimination: Cognitive Bias in Machine Learning
20mThere is no such thing as bias-free tech. How models shape our world.
38mSocial Implications of Bias in Machine Learning