Unmasking AI: My Mission to Protect What Is Human in a World of Machines by Joy Buolamwini.
Joy is a computer scientist and digital activist based at the MIT Media Lab. She founded the Algorithmic Justice League. She is a Rhodes Scholar, a Fulbright fellow, a Stamps Scholar, an Astronaut Scholar, and an Anita Borg Institute scholar. Joy holds a Masters and PhD in Media Arts & Sciences from MIT.
Unmasking AI is Joy’s memoir regarding her graduate work at Harvard. Joy is introducing her project regarding facial recognition. She realized her own face could not be identified unless she actually placed a white Halloween mask in-front of her face as a camera captured and attempted to identify her.
However, her work contributes to demonstrating the in accuracy of AI. Joy research is indeed like Dorthy in The Wizard of Oz, pulling back the curtain on the inaccuracies of AI facial recognition systems. And to some extent, also exposing the growing wealth of companies profiting from these already deployed.
Yet when informing her thesis committee that bias in AI has the potential to harm women and minorities, the perception includes many do not want to acknowledge these errors. Joy even receives feedback which questions the level of harm faulty facial recognition, and if is actually ‘significant’ enough to warrant this research.
AI Justice ?
So, Joy takes on the role of launching herself as the founder of The Algorithmic Justice League. Joy uses research, artwork, and policy to increase the awareness of artificial intelligence in society and addresses AI bias upon society.
I am reminded of Philip Delves Broughton‘s memoir of Harvard graduate school. As amazing as it may sound, Ahead of the Curve: Two Years at Harvard Business School was published in 2010. And also addresses the similar struggles of graduate students.
The interesting points along the way of her time include testifying before Congress, visiting the White House, and presenting at the World Economic Forum in Davos. In fact, her book culminates in the defense of her PhD thesis.
The book unravels the concept of the ‘coded gaze’, a term coined by Joy to illustrate biases programmed into AI systems. Her position is that these are not technical flaws but rather systemic inequalities.
In conclusion, Unmasking AI brings to light inaccuracies in AI systems. This a somewhat interesting story regarding her challenges at MIT. Yet, her research is indeed important to understand and hold companies providing these flawed systems accountable.