Yeah, But Should We....
summary
Software failures lead to concrete harm: MCAS causes Lion Air (189 dead) and Ethiopian (149 dead) crashes and regulators model 15 more crashes over 45 years; facial‑recognition systems from Amazon/Google misclassify women and people of color 10–100× more often, Clearview scrapes 3+ billion images and sells to 600+ law‑enforcement agencies, and a man receives eight years after a low‑confidence match. Historical examples show tabulating tech and punch cards enabled mass persecution, and state surveillance already imprisons Uyghurs at estimates of 120k–1M.
AI-written summary · about
chapters (10)
- 0:00Introduction and warning
- 3:31Software and its consequences
- 5:19Boeing 737 MAX disasters
- 13:34Developers behind the software
- 18:42Facial recognition bias
- 23:21IBM and the Holocaust
- 33:20IBM and apartheid
- 34:35Modern surveillance and oppression
- 38:11Our responsibility as developers
- 40:45Closing and resources
related talks
27mThe Limitations and Danger of Facial Recognition
01h 05mFacial recognition, a 'convenient' and 'efficient' solution looking for a problem?
45mThe Myth of Neutrality: How AI is widening social divides
54mFast and Loose with Facial Recognition Technology: Modern Tech, Race, and Civil Liberties
29mUnintended consequences of technology: Enabling NextGen discrimination
01h 15mHOW TO RECONCILE FACIAL RECOGNITION TECHNOLOGIES WITH CONSUMERS' PRIVACY
31mDoes This Algorithm Have a Soul?
34mDespicable machines: how computers can be assholes