Building a case for Fair and bias-free AI
summary
Argues that trustworthy AI requires measurable practices not slogans, listing five pillars — explainability, fairness, robustness, transparency, privacy — and urging ethics by design. Cites industry figures (three in four companies use AI; 78% of executives demand trust) and hard cases: an autonomous-vehicle fatality that sank a program and IBM halting facial recognition sales. Describes using Cloud Pak for Data to monitor models for drift, fairness and explanations in production, and reports that active monitoring uncovers and mitigates bias.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:50Building a case for fair and biasfree AI
- 2:47Ethics in AI
- 4:02Historical examples
- 4:18Food transparency
- 6:08What is AI
- 7:30AI in everyday business
- 10:18Pillars of AI ethics
- 11:40Trolley problem
- 12:28Autonomous vehicle problem
- 15:12Trustworthy AI
- 15:59Accountability
- 19:48Data Problems
- 21:12Value Alignment
- 22:00Explainability
- 26:06New ethical frameworks
- 29:18Privacy
- 30:12How long will it be used
- 32:28Principles
- 34:15Example
- 35:50Key message
- 36:06Conclusion
related talks
33mResponsible AI in Practice: Real-World Examples and Challenges
47mEthics & AI: Identifying & preventing bias in predictive models
40mEthical AI: Addressing Bias & Algorithmic Fairness
39mSponsored Session: The Search for Transparency and Accountability in the Age of AI: RAI & XAI as Essential Tools
26m(Un)ethical Artificial Intelligence: How to Keep the AI Fair for Everyone
11mAddressing Privacy, Bias, and Fairness in AI Systems- 39mThe Search for Transparency and Accountability in the Age of...
21mAlgorithms at Scale: Raising Awareness on Latent Inequities in Our Data