Ethical Machines Quotes

“Again, having diverse teams and consulting stakeholders is important and it ought to be done. But as a recent paper out of Columbia University found, they are not necessarily the most effective bias-identification and mitigation strategies.8 It’s more important, in the context of talking about bias mitigation in AI, that there exists expertise with regard to the ethical and legal risks that arise when training and testing your model.”
― Reid Blackman, Ethical Machines: Your Concise Guide to Totally Unbiased, Transparent, and Respectful AI

“Structure from Content Lesson #1: There must be an individual or, better, a set of individuals with relevant ethical, legal, and business expertise, who determine which, if any, of the mathematical metrics of fairness are appropriate for the particular use case.”
― Reid Blackman, Ethical Machines: Your Concise Guide to Totally Unbiased, Transparent, and Respectful AI

“The example brings the distinction between Structure and Content into stark relief. Structure concerns the way you identify and mitigate ethical risks. Content concerns what you take those ethical risks to be. An effective AI ethical risk program includes both.”
― Reid Blackman, Ethical Machines: Your Concise Guide to Totally Unbiased, Transparent, and Respectful AI