As artificial intelligence becomes increasingly essential to our lives, a steady stream of books is expressing delight or dismay. In this thoughtful addition to the genre, Mashable and Guardian contributing editor Havens warns that we have ignored an essential feature of machine thinking: its ethics.
In his previous book, Hacking Happiness: Why Your Personal Data Counts and How Tracking It Can Change the World (2014), the author explained how we might interact with the digital revolution to lead more fulfilling lives. This turns out to be difficult because we have little control over how today’s intrusive technology gathers information and none whatsoever about how it is used—mostly to sell stuff or amuse us. In the first half of the book, Havens delivers a dystopian account of where we may be heading. While AI developers “keep saying, ‘we need to make sure we understand the ethical issues around this technology,’ they nonetheless keep building systems they may not be able to control.” Should a driverless car on a crowded road swerve to miss a child who has run in front even if it means killing the driver? In most cases, these are situations that designers do not consider or consider incompletely. In the second half, Havens explains how to fix this. If we don’t understand our own ethics, we can’t program them into machines. Since most of us do not have a solid grasp of ethics, the author concentrates on identifying them and includes ingenious self-help exercises that prompt readers to discover what they truly value and then improve their lives by acting on those values (hint: altruism is more fulfilling than making money).
An astute philosophical meditation on a major problem facing designers of advanced computers.