As artificial intelligence (AI) is being increasingly used in many critical domains, one of the key challenges that we face in regulating AI is how to capture benefits of increased accuracy while ensuring that decisions are fair and non-discriminatory. What are the current legal approaches in addressing discrimination in the algorithmic setting and are they effective? Focusing on this question in the context of fair lending law, Professor Talia Gillis considers several regulatory approaches as reflected in recent proposals. In particular, she considers “input-based” approaches, which seek to scrutinize and police algorithmic inputs, and “human-in-the-loop" approaches, which require that human decision-makers retain decision-making authority when algorithms are deployed.
Lunch will be served on a first come, first served basis.
This lecture is made possible by the Clifford Chance Thought Leadership Initiative on Diversity at Columbia Law School.