AI as Employee: Managers Overlook Errors by 16%

Some companies are now treating AI systems as employees, giving them names and job titles. This shift, however, comes with an unexpected cost. New research indicates that managers review the work of AI less carefully when they perceive it as a colleague rather than just software. Here's how they found this out. Researchers surveyed about 1,200 managers, directors, and executives. Participants reviewed documents with errors. One group was told an AI tool produced the work. Another was told an AI employee named ALEX-3 created it. A third group believed a human employee named Alex wrote the documents. What's interesting is that for managers whose organizations already list AI agents on their org charts, treating AI as an employee reduced their error detection by about 16% compared to viewing it as a tool. This suggests that framing AI as an employee changes how managers delegate and take responsibility. This matters because it could lead to more mistakes going unnoticed in tasks performed by AI.