Man vs. Machine: Do the Means of Interaction Matter for Honesty?

Is dishonest behavior more prevalent in an interaction with a machine as opposed to a human? We analyze this question using an innovative experimental setup involving the reporting of an unobserved payout-relevant random draw either to an artificial conversational agent or a human. We find that reporting to an artificial conversational agent not able to demonstrate agency induces the lowest levels of honesty, whereas reporting to a human able to demonstrate such agency generates the highest levels of honest behavior. Our results have implications for designing efficient means of interaction between humans and machines in a variety of different contexts in the digital economy.