This research explores the potential dehumanizing effects of intelligent systems in the workplace, focusing on task and social deskilling. Leveraging Social Exchange Theory and Self-Determination Theory, the study investigates how interactions with AI impact employee competencies, social interactions and motivation over time. Using agent-based modeling in NetLogo, the study simulates workplace scenarios involving human and AI agents. The model aims to highlight trade-offs between productivity gains and maintaining employee well-being, emphasizing the need to design AI systems that complement rather than replace human interactions. Future work includes refining the model, validating its findings and analyzing emergent patterns to inform strategies for ethical AI integration in the workplace.