Take Action

Home | Faculty & Research Overview | Research

Research Details

A Simple Tactic That Could Help Reduce Bias in AI, Harvard Business Review

Abstract

It’s been well-established that AI-driven systems are subject to the biases of their human creators — we unwittingly “bake” biases into systems by training them on biased data or with “rules” created by experts with implicit biases. But it doesn’t have to be this way. The good news is that more strategic use of AI systems — through “blind taste tests” — can give us a fresh chance to identify and remove decision biases from the underlying algorithms even if we can’t remove them completely from our own habits of mind. That is, we can simply deny the algorithm the information suspected of biasing the outcome to ensure it makes predictions blind to that variable. Breaking the cycle of bias in this way has the potential to promote greater equality across contexts — from business to science to the arts — on dimensions including gender, race, socioeconomic status, and others.

Type

Article

Author(s)

Brian Uzzi

Date Published

2020

Citations

Uzzi, Brian. 2020. A Simple Tactic That Could Help Reduce Bias in AI. Harvard Business Review.

LINK
KELLOGG INSIGHT

Explore leading research and ideas

Find articles, podcast episodes, and videos that spark ideas in lifelong learners, and inspire those looking to advance in their careers.
learn more

COURSE CATALOG

Review Courses & Schedules

Access information about specific courses and their schedules by viewing the interactive course scheduler tool.
LEARN MORE

DEGREE PROGRAMS

Discover the path to your goals

Whether you choose our Full-Time, Part-Time or Executive MBA program, you’ll enjoy the same unparalleled education, exceptional faculty and distinctive culture.
learn more