Algorithmic Bias Tie-in for Martin Luther King Jr. Day

Rev. Dr. Martin Luther King Jr. marched so that his children would not be judged by the color of their skin... So perhaps looking at algorithms that identify people based on their physical characteristics isn't exactly what he had in mind. But the technology is there and the algorithms are judging people by their skin color, bone structure, etc... So I would argue that we can honor Rev. Dr. Martin Luther King Jr. by ensuring that all people are judged fairly and that people of color aren't misjudged by algorithms and the people who use rely on the algorithms (us).

This Better off Ted 'Racial Sensitivity' clip was shown at the virtual CSTA conference this summer. I remembered it today when thinking of how I could tie Martin Luther King Jr. Day into my CSP class content. I decided to start with this silly clip and ask the students these follow up questions:

  • What happened in this clip?
  • Who was effected by this "technology innovation"? How?
  • Can any of you relate to this experience--where technology wasn't made with you in mind?

 And then I continued the conversation with this more serious Ted Talk by Joy Juolamwini entitled, "How I'm Fighting Bias in Algorithms." (which I found in an older CS for All Teachers post by Bobby Oommen). I asked these follow up questions:

  • Joy says, "Who codes" matters. What does she mean?
  • How can we work to avoid algorithmic biases in our code?

The conversation flowed more naturally at that point, related to the beneficial and harmful effects of technology on society, economy, and culture. (I recognize we don't have the Explore PT anymore, but the concepts are still relevant for the AP Exam and life beyond the course).

Then, at the end of class, as a fun stretch into machine learning, I showed them this website:  And we considered abstractly how one might work with computers to create something like this.