Lecture(s)

Introduction (start here)

Interview with Ian Mercer

Here is Part I of my conversations with Ian Mercer recorded in June, 2020. The conversation is still relevant, so I have included it this class. In the introduction, I reference other interviews, so please disregard that comment.

I include these lectures because they are still relevant to our discussions today. Given advances in Machine Learning, advertising is going to become more sophisticated.

Part I: Ian Mercer (30 Minutes)

Part II Ian Mercer (42 minutes)

Terms and Subjects during the conversation with Ian

• The Robot Laws (my Sci-Fi fans should like this one)
• The Five V's of Data
• Right to be forgotten
• Fingerprinting (NYT 2019)
• M.L. = Machine Leanring
• Skynet

Read

• : "The Definition" (p9), and Introduction (10-24) and Chapter 3 (Behavioral Surplus)
• Chapter 4 from
• Big Mood Machine from The Baffler

Watch & Listen

Your Undivided Attention: Coded Bias with Joy Buolamwini

Bonus - Coded Bias

The film Coded Bias follows MIT Media Lab researcher Joy Buolamwini through her investigation of algorithmic discrimination after she accidentally discovers that facial recognition technologies do not detect darker-skinned faces. Joy is joined on screen by experts in the field, researchers, activists, and involuntary victims of algorithmic injustice.

www.humanetech.com

Shoshana Zuboff on Surveillance Capitalism

Case Study

You’re a product manager for a major social media firm. You’ve recently launched your first big project to millions of users: a service that connects people who share niche hobbies and post frequently on social media. However, you observe that most users appear to be white, from the North East united state, and female. You (and your advertisers) had hoped for a more diverse user segment.

Your Head of Engineering suggests applying a newly developed Machine Learning algorithm to determine the race, gender, and potential wealth of your users. This data would be invaluable in helping increase diversity on your platform by telling you who was using it. It would also become a treasure-trove of data for your advertisers, making the feature profitable.

What considerations would you have before using machine learning to glean socio-cultural insights from your users; and is there an ethical solution to your problem?

To Do