What we make matters: Ethics on the bleeding edge
summary
An Uber Volvo autonomous test vehicle kills a pedestrian in Tempe after the system identified the person six seconds out but had emergency braking disabled under computer control, relying on a distracted human watching diagnostics on an iPad off the sightline. Retailers infer sensitive states — Target claims 87% pregnancy prediction — and mobility data can reidentify people with four spatio‑temporal points at 95% accuracy. The talk argues to pick a concrete ethical framework (consequentialist, duty, virtue) before writing code that places unrealistic expectations on users.
AI-written summary · about
chapters (10)
- 0:00Start
- 0:16WHAT WE MAKE MATTERS: ETHICS ON THE BLEEDING EDGE
- 3:10WHAT HAPPENED HERE?
- 7:44THE TROLLEY PROBLEM
- 18:05BUT DANIEL... I WRITE JAVASCRIPT.
- 19:5825 MARCH 2013 SCIENTIFIC REPORTS
- 24:00QUIZ TIME!
- 24:38THE CONSEQUENTIALIST VIEW
- 25:13THE DUTY VIEW
- 25:38THE VIRTUE VIEW
related talks
15mUtility vs Sanctity: Designing an ethically acceptable AI
45mFuture Ethics
47mThe Ethics of Software - some practical considerations
10mSara Holoubek – All Tech Is Human 2018
41mMachine Ethics and Emerging Technologies
41mBeing Good: An Introduction to Robo- and Machine Ethics
16mExistential Engineering or How Ethics Precedes Error Handling
17mEthics and the Cost of Moral Tech Debt