Keynote: Ethics in AI: Building Fair and Just Intelligence Systems
summary
Shows concrete evidence and actionable methods: pulse oximeters overestimate oxygenation for Black patients leading to worse outcomes, a resume‑screener trained on past hires penalizes women, a video‑interview system denies a deaf indigenous applicant, and NSW STOMP disproportionately targets Indigenous youth. Presents a practical three‑phase workflow — before/while/after — with tools like Fairlearn AI F3 WhatIf plus operational steps: test across demographics, add monitoring dashboards and appeals, compensate community experts, and plan model retirement.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:49AI's impact on daily life
- 4:16Ethical AI premise
- 7:28AI bias in health
- 10:15Bias in workplace AI
- 12:25AI bias in justice system
- 14:17Bias sources in AI
- 17:51Historical bias in models
- 20:01Bias after deployment
- 21:47AI's societal impact
- 22:50Anti-oppressive AI frameworks
- 27:08Steps to reduce AI bias
- 29:43Actions for AI ethics
- 32:12Future of AI responsibility
- 33:20Closing call to action
related talks
37mKeynote: Ethical Machine Learning: Creating Fair Models in an Unjust World
26m(Un)ethical Artificial Intelligence: How to Keep the AI Fair for Everyone
40mEthical AI: Addressing Bias & Algorithmic Fairness
30mFireside chat: Breaking the wall of unethical AI
47mEthics & AI: Identifying & preventing bias in predictive models
59mIn search of inclusive AI
05mToward Ethical AI
40mResponsible AI Practices: Fairness in ML