Wharton Study Reveals Deep Uncertainty Over AI Agent Accountability in the Workplace

Managers are giving AI agents greater autonomy at work, but a new Wharton study finds that organizations struggle to assign clear responsibility when these autonomous systems fail. When asked who should be held accountable for consequential errors made by AI agents, managers split almost evenly: 28% named a governance or risk team, 26% pointed to the deploying employee's manager, and 25% blamed the employee themselves. Roughly one in five managers or directors said nobody was clearly accountable or they did not know.