Ghostwritten Feedback: Managers' Hidden Use of AI in Performance Conversations, and How Leaders Can Protect Trust and Development

Abstract: Generative AI has moved quietly into one of the most personal tasks in management: writing feedback and performance reviews. A 2026 survey of more than 1,000 U.S. corporate professionals found that 78% of managers had used AI for feedback or reviews in the prior year, yet only 16% of individual contributors had been told that AI played a role in their latest review. This article examines what undisclosed, AI-assisted feedback does to trust, feedback quality, fairness, and employee development. Drawing on research on AI disclosure, AI-mediated communication, feedback intervention theory, organizational justice, automation bias, and gendered review language, it argues that the central risk is less the tool than the substitution of machine fluency for managerial judgment. Five evidence-based responses are outlined: disclosure norms, human ownership of evaluative judgment, AI for preparation and rehearsal, bias calibration, and governance aligned with emerging regulation such as the EU AI Act. Organizational examples span technology, finance, logistics, platform work, and government.