April 9, 2020

Probably approximately ethical


  • When it comes to applying AI to a business, there are many ethical tradeoffs to consider (especially in highly regulated industries like finance and healthcare).

Artificial intelligence is so powerful because it can automate tasks in use cases where precise rules are hard to describe (such as calculating the likelihood of a customer repaying their bill). But this brings up many considerations for an organization, including questions of accuracy, error, explainability, privacy, and fairness.

In this video, senior director of product and business development at Borealis AI Kathryn Hume presents three rules of thumb for managing the risks that arise when applying machine learning systems to real-world applications.

“When we move from rules to systems that make approximated guesses, you can imagine that there might be implications for an enterprise.”
This tech talk was part of the Harvard Business School Executive Education Advanced Management Program.

Related Topics

People in this Article