When AI Joins the Team: Three Principles for Responsible Agent Design
Published in Microsoft Design
Authors: Hilary Dwyer, Jeffrey Basoah, Jennifer Wang
As AI agents increasingly join enterprise work environments, this article presents three core principles for designing them responsibly. The authors argue that AI agents should be differentiated from humans through clear naming and visual markers, built for intent by adapting to specific workplace contexts rather than adding unnecessary personality, and made bias resistant by disclosing limitations and preventing over-reliance. The piece emphasizes that effective workplace AI agents should prioritize transparency, trustworthiness, and social awareness over entertainment value.
Recommended citation: Dwyer, H., Basoah, J., and Wang, J. 2026. When AI Joins the Team: Three Principles for Responsible Agent Design. Microsoft Design. https://microsoft.design/articles/principles-for-responsible-agent-design/