The AI Transparency Paradox

Author: Andrew Burt

Publisher: Harvard Business Review

Publication Year: 2019

Summary: The following article describes how there are risks to achieving AI transparency. First, corporations often expose themselves to financial risk by publicizing what they are doing, since this could give rise to lawsuits. Second, there is a risk that bad actors can reverse engineer algorithms based on their public explanations. This is a financial risk, in that competitors might be able to improve their own technology via disclosure. However, it’s also a social risk because bad actors could, for example, use this information to compromise the privacy of those whose data was used for modeling.