Business Ethics in the Age of Robotics Lecture Features Wharton Business Professor


Professor Thomas Donaldson of the University of Pennsylvania Wharton School of Business gave a lecture surrounding the importance of intrinsic values in business strategy and ethical considerations in the age of robotics on Monday in the Law School Library.

Professor Thomas Donaldson

Donaldson, in collaboration with Business Administration Professor James Walsh of the University of Michigan, is synthesizing a new network of value conceptions with which to approach management theory. This “topology” distinguishes intrinsic values such as happiness and fairness from non-intrinsic ones, unlike traditional management theory.

“Contemporary theories of business strategy and value creation use mostly non-intrinsic value concepts … they work well with problems if you’re talking just about overall strategy, but they work less well at the level of business systems and their impact on humanity, on society,” Donaldson said. “We need another map.”

The Wharton professor defines an intrinsic value as the end to the chain of reasons that justify our actions; that is, we go to class to get a good grade, to graduate with a high GPA, to secure a good job, to achieve financial independence, to be happy overall. Happiness is the intrinsic value.

He tied this notion of intrinsic value consideration in business to contemporary business trends, including the emergence of autonomous vehicles.

He framed this discussion in the context of a modern take on the classic trolley problem: a self-driving car can “choose” between swerving to avoid killing four pedestrians, in which case the driver would die, or not swerving, in which case the driver would survive but the pedestrians would die.

“It’s an instance of a problem that more sophisticated engineering cannot solve, and that requires a more sophisticated appeal to values,” Donaldson said. “We’re starting to work at a level where we’re going to have to go a little deeper with what constitutes intrinsic worth.”

Donaldson referenced economist Alan Greenspan’s new book, The Map and the Territory, to make the point that maps are most effective when they highlight particular elements, such as topology or political boundaries, as opposed to displaying all possible information about a territory. He related this concept to the notion of decision making artificial intelligence.

“AI [artificial intelligence] is extremely good at working with a single map, but as soon as the question becomes, ‘Do we need to put this map down because we need the map that shows the topology instead of the state boundaries?’ handling that kind of problem turns out to be extremely difficult even for the most sophisticated computers we have.”

Mercedes-Benz has rather famously answered the question of the trolley problem by stating they will program their cars to save the driver every time. Christoph von Hugo, the automaker’s manager of Driver Assistance Systems and Active Safety said, “If you know you can save at least one person, at least save that one. Save the one in the car.”

Donaldson, and indeed most moral philosophers, would agree that this is too simple a solution to a decades-old problem.

“Such high-level analysis of intertwined intrinsic values is predictably challenging, if not impossible, for any robot or AI system imaginable,” Donaldson said. “The bottom line is, the maps that we have are not up to the task of some of the challenges we have … modernity is putting special challenges in our face. It’s time we shifted our thinking.”