Eye on AI: Decoding the Bias
In recent years the application of artificial intelligence (AI) in government, business and greater civil society has grown. As AI use has become widespread it has created some deep ethical oversights. Left unchecked and without proper controls, AI and automated systems can perpetrate racial, gender, and other forms of bias that should concern everyone.
To address these issues, TPL partnered with MIT libraries for the program Eye on AI: Decoding the Bias. The program premiered on Tuesday, January 25. If you missed it, you can watch the full recording on Crowdcast.
The discussion centered on Shalini Kantayya's 2020 documentary Coded Bias. The documentary tells the story of Joy Buolamwini, a Black American who finds out that a facial recognition program recognizes her face only with a white mask on. This spurs her to investigate how AI algorithms affect minorities. The documentary interweaves stories of racial and gender minorities harmed by AI. These include wrongful racial profiling by facial recognition systems, assignment of higher reoffending scores to minorities, exclusion of healthcare coverage, and discriminatory applicant tracking systems in jobs and education.
Algorithms of Oppression by Safiya Noble
The program featured a thoughtful discussion of the documentary Coded Bias. Speakers were:
- Shalini Kantayya (director of Coded Bias), Safiya Noble (author of Algorithms of Oppression)
- Kishana Peck (founder and executive director of Toronto Womxn in Data Science)
- Sasha Constana-Shock (director of Research & Design at Algorithmic Justice League)
- Ethan Zuckerman (associate professor of public policy at University of Massachusetts at Amherst)
Some major takeaways were actionable proposals to hold both corporations and governments accountable. These include interventions in the classroom to educate students across education levels, as well as, interventions in design space to ensure corporate actors, designers and engineers create systems that don't perpetuate inequality. And also, legal interventions to hold states accountable. A positive consequence of this activism recently led to the ban on government use of facial recognition in the US.
More books on AI and algorithmic bias
AI Ethics by Mark Coeckelbergh
Ethics of Artificial Intelligence edited by Matthew Liao
More helpful resources
Here are some videos on algorithmic bias and the fundamentals of ethics in AI. You will need a library card to watch them.
If you have thoughts or experiences to share please leave a comment below.
Comments